diff --git "a/awesome-japanese-nlp-resources-search.json" "b/awesome-japanese-nlp-resources-search.json" --- "a/awesome-japanese-nlp-resources-search.json" +++ "b/awesome-japanese-nlp-resources-search.json" @@ -4375,6 +4375,22 @@ "source": "GitHub", "score": 0.0 }, + { + "description": "Dataset for the LREC-COLING 2024 paper \"A Gaze-grounded Visual Question Answering Dataset for Clarifying Ambiguous Japanese Questions\"", + "url": "https://github.com/riken-grp/GazeVQA", + "project_name": "GazeVQA", + "stargazers_count": 0, + "source": "GitHub", + "score": 0.0 + }, + { + "description": "Code for J-CRe3 experiments (Ueda et al., LREC-COLING, 2024)", + "url": "https://github.com/riken-grp/J-CRe3", + "project_name": "J-CRe3", + "stargazers_count": 0, + "source": "GitHub", + "score": 0.0 + }, { "description": "spaCy tutorial in English and Japanese. spacy-transformers, BERT, GiNZA.", "url": "https://github.com/yuibi/spacy_tutorial", @@ -4559,5620 +4575,5540 @@ "source": "GitHub", "score": 0.0 }, - { - "url": "https://huggingface.co/tohoku-nlp/bert-base-japanese", - "downloads": 1338802, - "description": "BERT base Japanese (IPA dictionary)This is a BERT model pretrained on texts in the Japanese language.", - "source": "Hugging Face", - "score": 0.1860571575962085, - "project_name": "bert-base-japanese" - }, { "url": "https://huggingface.co/jonatasgrosman/wav2vec2-large-xlsr-53-japanese", - "downloads": 1280052, + "downloads": 1319775, "description": "Fine-tuned XLSR-53 large model for speech recognition in JapaneseFine-tuned facebook/wav2vec2-large-xlsr-53 on Japanese using the train and validation splits of Common Voice 6.1, CSS10 and JSUT.When using this model, make sure that your speech input is sampled at 16kHz.", "source": "Hugging Face", - "score": 0.17789250142690397, + "score": 0.19746723056164767, "project_name": "wav2vec2-large-xlsr-53-japanese" }, + { + "url": "https://huggingface.co/tohoku-nlp/bert-base-japanese", + "downloads": 1179147, + "description": "BERT base Japanese (IPA dictionary)This is a BERT model pretrained on texts in the Japanese language.", + "source": "Hugging Face", + "score": 0.1764262033415356, + "project_name": "bert-base-japanese" + }, { "url": "https://huggingface.co/tsmatz/xlm-roberta-ner-japanese", - "downloads": 1073743, + "downloads": 1053882, "description": "xlm-roberta-ner-japanese(Japanese caption : 日本語の固有表現抽出のモデル)This model is a fine-tuned version of xlm-roberta-base (pre-trained cross-lingual RobertaModel) trained for named entity recognition (NER) token classification.", "source": "Hugging Face", - "score": 0.14922114739059675, + "score": 0.15768381722548946, "project_name": "xlm-roberta-ner-japanese" }, { "url": "https://huggingface.co/kha-white/manga-ocr-base", - "downloads": 752737, + "downloads": 754295, "description": "Manga OCROptical character recognition for Japanese text, with the main focus being Japanese manga.", "source": "Hugging Face", - "score": 0.10461002197300062, + "score": 0.1128590439101347, "project_name": "manga-ocr-base" }, - { - "url": "https://huggingface.co/sonoisa/sentence-bert-base-ja-mean-tokens-v2", - "downloads": 471876, - "description": "This is a Japanese sentence-BERT model.", - "source": "Hugging Face", - "score": 0.06557796246037015, - "project_name": "sentence-bert-base-ja-mean-tokens-v2" - }, { "url": "https://huggingface.co/tohoku-nlp/bert-base-japanese-whole-word-masking", - "downloads": 385120, + "downloads": 329509, "description": "BERT base Japanese (IPA dictionary, whole word masking enabled)This is a BERT model pretrained on texts in the Japanese language.", "source": "Hugging Face", - "score": 0.05352123206676702, + "score": 0.04930175952350814, "project_name": "bert-base-japanese-whole-word-masking" }, + { + "url": "https://huggingface.co/sonoisa/sentence-bert-base-ja-mean-tokens-v2", + "downloads": 233964, + "description": "This is a Japanese sentence-BERT model.", + "source": "Hugging Face", + "score": 0.035006135993730245, + "project_name": "sentence-bert-base-ja-mean-tokens-v2" + }, { "url": "https://huggingface.co/tohoku-nlp/bert-base-japanese-char", - "downloads": 133281, + "downloads": 130635, "description": "BERT base Japanese (character tokenization)This is a BERT model pretrained on texts in the Japanese language.", "source": "Hugging Face", - "score": 0.01852244321533749, + "score": 0.01954585566814104, "project_name": "bert-base-japanese-char" }, { "url": "https://huggingface.co/tohoku-nlp/bert-base-japanese-char-v2", - "downloads": 113466, + "downloads": 111700, "description": "BERT base Japanese (character-level tokenization with whole word masking, jawiki-20200831)This is a BERT model pretrained on texts in the Japanese language.", "source": "Hugging Face", - "score": 0.015768695777128652, + "score": 0.016712765171136023, "project_name": "bert-base-japanese-char-v2" }, - { - "url": "https://huggingface.co/tohoku-nlp/bert-base-japanese-char-whole-word-masking", - "downloads": 109652, - "description": "BERT base Japanese (character tokenization, whole word masking enabled)This is a BERT model pretrained on texts in the Japanese language.", - "source": "Hugging Face", - "score": 0.01523865324726095, - "project_name": "bert-base-japanese-char-whole-word-masking" - }, { "url": "https://huggingface.co/sonoisa/sentence-luke-japanese-base-lite", - "downloads": 97584, + "downloads": 92168, "description": "This is a Japanese sentence-LUKE model.", "source": "Hugging Face", - "score": 0.013561528640432573, + "score": 0.013790350405490285, "project_name": "sentence-luke-japanese-base-lite" }, { - "url": "https://huggingface.co/elyza/ELYZA-japanese-Llama-2-7b-instruct", - "downloads": 94112, - "description": "ELYZA-japanese-Llama-2-7bModel DescriptionELYZA-japanese-Llama-2-7b", + "url": "https://huggingface.co/tohoku-nlp/bert-base-japanese-char-whole-word-masking", + "downloads": 77576, + "description": "BERT base Japanese (character tokenization, whole word masking enabled)This is a BERT model pretrained on texts in the Japanese language.", "source": "Hugging Face", - "score": 0.013079014832435546, - "project_name": "ELYZA-japanese-Llama-2-7b-instruct" + "score": 0.01160706777901565, + "project_name": "bert-base-japanese-char-whole-word-masking" }, { - "url": "https://huggingface.co/hotchpotch/japanese-reranker-cross-encoder-xsmall-v1", - "downloads": 85977, - "description": "hotchpotch/japanese-reranker-cross-encoder-xsmall-v1日本語で学習させた Reranker (CrossEncoder) シリーズです。", + "url": "https://huggingface.co/elyza/ELYZA-japanese-Llama-2-7b-instruct", + "downloads": 75195, + "description": "ELYZA-japanese-Llama-2-7bModel DescriptionELYZA-japanese-Llama-2-7b", "source": "Hugging Face", - "score": 0.01194847052711993, - "project_name": "japanese-reranker-cross-encoder-xsmall-v1" + "score": 0.01125081805768642, + "project_name": "ELYZA-japanese-Llama-2-7b-instruct" }, { "url": "https://huggingface.co/setu4993/LaBSE", - "downloads": 76287, + "downloads": 71711, "description": "LaBSEModel descriptionLanguage-agnostic BERT Sentence Encoder (LaBSE) is a BERT-based model trained for sentence embedding for 109 languages.", "source": "Hugging Face", - "score": 0.010601823407450807, + "score": 0.010729535391113118, "project_name": "LaBSE" }, { "url": "https://huggingface.co/pkshatech/GLuCoSE-base-ja", - "downloads": 69039, + "downloads": 68020, "description": "GLuCoSE (General Luke-based Contrastive Sentence Embedding)-base-Japanese日本語のREADME/Japanese READMEGLuCoSE (General LUke-based COntrastive Sentence Embedding, \"glucose\") is a Japanese text embedding model based on LUKE.", "source": "Hugging Face", - "score": 0.00959454803868282, + "score": 0.010177280993202078, "project_name": "GLuCoSE-base-ja" }, - { - "url": "https://huggingface.co/tohoku-nlp/bert-base-japanese-v3", - "downloads": 66491, - "description": "BERT base Japanese (unidic-lite with whole word masking, CC-100 and jawiki-20230102)This is a BERT model pretrained on texts in the Japanese language.", - "source": "Hugging Face", - "score": 0.009240445163459196, - "project_name": "bert-base-japanese-v3" - }, { "url": "https://huggingface.co/sonoisa/sentence-bert-base-ja-mean-tokens", - "downloads": 64860, + "downloads": 65208, "description": "This is a Japanese sentence-BERT model.", "source": "Hugging Face", - "score": 0.009013780410912206, + "score": 0.00975654423705853, "project_name": "sentence-bert-base-ja-mean-tokens" }, { "url": "https://huggingface.co/colorfulscoop/sbert-base-ja", - "downloads": 62766, + "downloads": 62127, "description": "Sentence BERT base Japanese modelThis repository contains a Sentence BERT base model for Japanese.", "source": "Hugging Face", - "score": 0.008722771219107547, + "score": 0.009295559192364975, "project_name": "sbert-base-ja" }, + { + "url": "https://huggingface.co/tohoku-nlp/bert-base-japanese-v3", + "downloads": 59856, + "description": "BERT base Japanese (unidic-lite with whole word masking, CC-100 and jawiki-20230102)This is a BERT model pretrained on texts in the Japanese language.", + "source": "Hugging Face", + "score": 0.008955767878992996, + "project_name": "bert-base-japanese-v3" + }, + { + "url": "https://huggingface.co/hotchpotch/japanese-reranker-cross-encoder-xsmall-v1", + "downloads": 58418, + "description": "hotchpotch/japanese-reranker-cross-encoder-xsmall-v1日本語で学習させた Reranker (CrossEncoder) シリーズです。", + "source": "Hugging Face", + "score": 0.008740611600424567, + "project_name": "japanese-reranker-cross-encoder-xsmall-v1" + }, { "url": "https://huggingface.co/llm-book/bert-base-japanese-v3-ner-wikipedia-dataset", - "downloads": 40662, + "downloads": 40034, "description": "llm-book/bert-base-japanese-v3-ner-wikipedia-dataset「大規模言語モデル入門」の第6章で紹介している固有表現認識のモデルです。", "source": "Hugging Face", - "score": 0.005650914879255506, + "score": 0.005989962765096325, "project_name": "bert-base-japanese-v3-ner-wikipedia-dataset" }, { "url": "https://huggingface.co/ku-nlp/deberta-v2-base-japanese", - "downloads": 39876, + "downloads": 38842, "description": "Model Card for Japanese DeBERTa V2 baseModel", "source": "Hugging Face", - "score": 0.0055416822026755335, + "score": 0.00581161347159593, "project_name": "deberta-v2-base-japanese" }, { "url": "https://huggingface.co/sazyou-roukaku/BracingEvoMix", - "downloads": 37747, + "downloads": 38390, "description": "License:CreativeML Open RAIL-MAdditional Copyright: sazyou_roukaku (TwitterID @sazyou_roukaku) as of May 31, 2023このモデルは『CreativeML Open RAIL-M』でLicenseそのものに変更はありません。", "source": "Hugging Face", - "score": 0.005245808960387034, + "score": 0.00574398437708068, "project_name": "BracingEvoMix" }, { "url": "https://huggingface.co/cyberagent/open-calm-3b", - "downloads": 35042, + "downloads": 37050, "description": "OpenCALM-3BModel DescriptionOpenCALM is a suite of decoder-only language models pre-trained on Japanese datasets, developed by CyberAgent, Inc.", "source": "Hugging Face", - "score": 0.004869887344421608, + "score": 0.005543491043783255, "project_name": "open-calm-3b" }, { "url": "https://huggingface.co/rinna/japanese-gpt-neox-3.6b", - "downloads": 32238, + "downloads": 36125, "description": "japanese-gpt-neox-3.6bOverviewThis repository provides a Japanese GPT-NeoX model of 3.6 billion parameters.", "source": "Hugging Face", - "score": 0.004480207414230461, + "score": 0.005405090795051824, "project_name": "japanese-gpt-neox-3.6b" }, { "url": "https://huggingface.co/rinna/japanese-clip-vit-b-16", - "downloads": 25908, + "downloads": 27728, "description": "rinna/japanese-clip-vit-b-16This is a Japanese CLIP (Contrastive Language-Image Pre-Training) model trained by rinna Co.", "source": "Hugging Face", - "score": 0.0036005091410100743, + "score": 0.004148715780351473, "project_name": "japanese-clip-vit-b-16" }, - { - "url": "https://huggingface.co/tohoku-nlp/bert-base-japanese-v2", - "downloads": 20502, - "description": "BERT base Japanese (unidic-lite with whole word masking, jawiki-20200831)This is a BERT model pretrained on texts in the Japanese language.", - "source": "Hugging Face", - "score": 0.00284922180056309, - "project_name": "bert-base-japanese-v2" - }, { "url": "https://huggingface.co/augmxnt/shisa-gamma-7b-v1", - "downloads": 19648, + "downloads": 20458, "description": "shisa-gamma-7b-v1For more information see our main Shisa 7B modelWe applied a version of our fine-tune data set onto Japanese Stable LM Base Gamma 7B and it performed pretty well, just sharing since it might be of interest.", "source": "Hugging Face", - "score": 0.0027305389687573697, + "score": 0.003060964636267688, "project_name": "shisa-gamma-7b-v1" }, + { + "url": "https://huggingface.co/ReadyON/karakuri-lm-8x7b-instruct-v0.1-gguf", + "downloads": 19800, + "description": "KARAKURI LM 8x7B Instruct v0.1 GGUFGGUF version of KARAKURI LM 8x7B Instruct v0.1Debeloped by: KARAKURI Inc.Languages: Primarily English and JapaneseLicense: Apache 2.0Finetuned from model: tokyotech-llm/Swallow-MX-8x7b-NVE-v0.1", + "source": "Hugging Face", + "score": 0.0029625134323052215, + "project_name": "karakuri-lm-8x7b-instruct-v0.1-gguf" + }, + { + "url": "https://huggingface.co/tohoku-nlp/bert-base-japanese-v2", + "downloads": 19621, + "description": "BERT base Japanese (unidic-lite with whole word masking, jawiki-20200831)This is a BERT model pretrained on texts in the Japanese language.", + "source": "Hugging Face", + "score": 0.002935731113902058, + "project_name": "bert-base-japanese-v2" + }, { "url": "https://huggingface.co/tokyotech-llm/Swallow-7b-NVE-instruct-hf", - "downloads": 18708, + "downloads": 17682, "description": "SwallowOur Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data.", "source": "Hugging Face", - "score": 0.002599904470048497, + "score": 0.002645614268182875, "project_name": "Swallow-7b-NVE-instruct-hf" }, { "url": "https://huggingface.co/rinna/japanese-gpt-neox-3.6b-instruction-sft-v2", - "downloads": 17338, + "downloads": 17479, "description": "japanese-gpt-neox-3.6b-instruction-sft-v2OverviewThis repository provides a Japanese GPT-NeoX model of 3.6 billion parameters.", "source": "Hugging Face", - "score": 0.002409511636823864, + "score": 0.0026152410244072205, "project_name": "japanese-gpt-neox-3.6b-instruction-sft-v2" }, { - "url": "https://huggingface.co/KoichiYasuoka/bert-base-japanese-upos", - "downloads": 13073, - "description": "bert-base-japanese-uposModel DescriptionThis is a BERT model pre-trained on Japanese Wikipedia texts for POS-tagging and dependency-parsing, derived from bert-base-japanese-char-extended.", + "url": "https://huggingface.co/stabilityai/japanese-stablelm-base-gamma-7b", + "downloads": 13538, + "description": "Japanese Stable LM Base Gamma 7BModel", "source": "Hugging Face", - "score": 0.0018167923421500965, - "project_name": "bert-base-japanese-upos" + "score": 0.002025581153866065, + "project_name": "japanese-stablelm-base-gamma-7b" }, { "url": "https://huggingface.co/rinna/llama-3-youko-8b", - "downloads": 11675, + "downloads": 12388, "description": "Llama 3 Youko 8B (rinna/llama-3-youko-8b)", "source": "Hugging Face", - "score": 0.0016225082685383902, + "score": 0.0018535159797675295, "project_name": "llama-3-youko-8b" }, - { - "url": "https://huggingface.co/stabilityai/japanese-stablelm-base-gamma-7b", - "downloads": 11324, - "description": "Japanese Stable LM Base Gamma 7BModel", - "source": "Hugging Face", - "score": 0.0015737287908290135, - "project_name": "japanese-stablelm-base-gamma-7b" - }, { "url": "https://huggingface.co/elyza/ELYZA-japanese-Llama-2-7b-fast", - "downloads": 10299, + "downloads": 11753, "description": "ELYZA-japanese-Llama-2-7bModel DescriptionELYZA-japanese-Llama-2-7b", "source": "Hugging Face", - "score": 0.0014312815980879556, + "score": 0.0017585060792870337, "project_name": "ELYZA-japanese-Llama-2-7b-fast" }, { - "url": "https://huggingface.co/mmnga/Phi-3-medium-128k-instruct-gguf", - "downloads": 10049, - "description": "Phi-3-medium-128k-instruct-ggufmicrosoftさんが公開しているPhi-3-medium-128k-instructのggufフォーマット変換版です。", + "url": "https://huggingface.co/KoichiYasuoka/bert-base-japanese-upos", + "downloads": 10792, + "description": "bert-base-japanese-uposModel DescriptionThis is a BERT model pre-trained on Japanese Wikipedia texts for POS-tagging and dependency-parsing, derived from bert-base-japanese-char-extended.", "source": "Hugging Face", - "score": 0.0013965383803462342, - "project_name": "Phi-3-medium-128k-instruct-gguf" + "score": 0.0016147194424968661, + "project_name": "bert-base-japanese-upos" }, { "url": "https://huggingface.co/sazyou-roukaku/chilled_remix", - "downloads": 9909, + "downloads": 8653, "description": "【告知】chilled_remix及びreversemixは2023年5月21日にVersion変更を行い、v2へ移行いたしました。", "source": "Hugging Face", - "score": 0.0013770821784108702, + "score": 0.00129467821867359, "project_name": "chilled_remix" }, + { + "url": "https://huggingface.co/bclavie/JaColBERT", + "downloads": 8557, + "description": "このドキュメントの日本語版はまだ作成中です。", + "source": "Hugging Face", + "score": 0.0012803145171836252, + "project_name": "JaColBERT" + }, + { + "url": "https://huggingface.co/tohoku-nlp/bert-large-japanese-v2", + "downloads": 8529, + "description": "BERT large Japanese (unidic-lite with whole word masking, CC-100 and jawiki-20230102)This is a BERT model pretrained on texts in the Japanese language.", + "source": "Hugging Face", + "score": 0.0012761251042490522, + "project_name": "bert-large-japanese-v2" + }, + { + "url": "https://huggingface.co/sociocom/MedNERN-CR-JA", + "downloads": 8526, + "description": "This is a model for named entity recognition of Japanese medical documents.", + "source": "Hugging Face", + "score": 0.0012756762385774908, + "project_name": "MedNERN-CR-JA" + }, { "url": "https://huggingface.co/mmnga/stockmark-gpt-neox-japanese-1.4b-gguf", - "downloads": 8818, + "downloads": 8381, "description": "stockmark-gpt-neox-japanese-1.4b-ggufstockmarkさんが公開しているgpt-neox-japanese-1.4bのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 0.001225462776185998, + "score": 0.0012539810644520233, "project_name": "stockmark-gpt-neox-japanese-1.4b-gguf" }, { "url": "https://huggingface.co/mmnga/Phi-3-mini-128k-instruct-gguf", - "downloads": 8488, + "downloads": 8183, "description": "Phi-3-mini-128k-instruct-ggufmicrosoftさんが公開しているPhi-3-mini-128k-instructのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 0.0011796017287669257, + "score": 0.0012243559301289711, "project_name": "Phi-3-mini-128k-instruct-gguf" }, { - "url": "https://huggingface.co/tohoku-nlp/bert-large-japanese-v2", - "downloads": 8369, - "description": "BERT large Japanese (unidic-lite with whole word masking, CC-100 and jawiki-20230102)This is a BERT model pretrained on texts in the Japanese language.", - "source": "Hugging Face", - "score": 0.0011630639571218663, - "project_name": "bert-large-japanese-v2" - }, - { - "url": "https://huggingface.co/abeja/gpt-neox-japanese-2.7b", - "downloads": 7186, - "description": "gpt-neox-japanese-2.7bThe open PR is merged on 2022/9/14.You can use this model with v4.23 and higher versions of transformers as follows,pip install transformersThis repository provides a 2.7B-parameter Japanese GPT-NeoX-based model.", + "url": "https://huggingface.co/QuantFactory/llama-3-youko-8b-GGUF", + "downloads": 7711, + "description": "QuantFactory/llama-3-youko-8b-GGUFThis is quantized version of rinna/llama-3-youko-8b created using llama.cppModel DescriptionOverviewWe conduct continual pre-training of meta-llama/Meta-Llama-3-8B on 22B tokens from a mixture of Japanese and English datasets.", "source": "Hugging Face", - "score": 0.0009986590507680404, - "project_name": "gpt-neox-japanese-2.7b" + "score": 0.0011537343978033114, + "project_name": "llama-3-youko-8b-GGUF" }, { "url": "https://huggingface.co/rinna/japanese-gpt2-medium", - "downloads": 6724, + "downloads": 7462, "description": "japanese-gpt2-mediumThis repository provides a medium-sized Japanese GPT-2 model.", "source": "Hugging Face", - "score": 0.0009344535843813393, + "score": 0.0011164785470637153, "project_name": "japanese-gpt2-medium" }, { - "url": "https://huggingface.co/rinna/japanese-roberta-base", - "downloads": 6708, - "description": "japanese-roberta-baseThis repository provides a base-sized Japanese RoBERTa model.", + "url": "https://huggingface.co/Aratako/Oumuamua-7b-RP-GGUF", + "downloads": 7336, + "description": "Ninja-v1-RP-expressive-GGUF概要Aratako/Oumuamua-7b-RPの量子化済みGGUF版です。", "source": "Hugging Face", - "score": 0.0009322300184458691, - "project_name": "japanese-roberta-base" + "score": 0.0010976261888581366, + "project_name": "Oumuamua-7b-RP-GGUF" + }, + { + "url": "https://huggingface.co/tokyotech-llm/Swallow-7b-instruct-v0.1", + "downloads": 7255, + "description": "SwallowOur Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data.", + "source": "Hugging Face", + "score": 0.001085506815725979, + "project_name": "Swallow-7b-instruct-v0.1" + }, + { + "url": "https://huggingface.co/line-corporation/line-distilbert-base-japanese", + "downloads": 7130, + "description": "LINE DistilBERT", + "source": "Hugging Face", + "score": 0.0010668040794109207, + "project_name": "line-distilbert-base-japanese" + }, + { + "url": "https://huggingface.co/abeja/gpt-neox-japanese-2.7b", + "downloads": 7044, + "description": "gpt-neox-japanese-2.7bThe open PR is merged on 2022/9/14.You can use this model with v4.23 and higher versions of transformers as follows,pip install transformersThis repository provides a 2.7B-parameter Japanese GPT-NeoX-based model.", + "source": "Hugging Face", + "score": 0.0010539365968261607, + "project_name": "gpt-neox-japanese-2.7b" + }, + { + "url": "https://huggingface.co/mmnga/Phi-3-medium-128k-instruct-gguf", + "downloads": 6790, + "description": "Phi-3-medium-128k-instruct-ggufmicrosoftさんが公開しているPhi-3-medium-128k-instructのggufフォーマット変換版です。", + "source": "Hugging Face", + "score": 0.0010159326366339623, + "project_name": "Phi-3-medium-128k-instruct-gguf" }, { "url": "https://huggingface.co/sonoisa/t5-base-japanese", - "downloads": 6523, + "downloads": 6562, "description": "日本語T5事前学習済みモデルThis is a T5 (Text-to-Text Transfer Transformer) model pretrained on Japanese corpus.", "source": "Hugging Face", - "score": 0.0009065200373169953, + "score": 0.0009818188455952962, "project_name": "t5-base-japanese" }, { "url": "https://huggingface.co/mmnga/ArrowPro-7B-KillerWhale-gguf", - "downloads": 6436, + "downloads": 6558, "description": "ArrowPro-7B-KillerWhale-ggufDataPilotさんが公開しているArrowPro-7B-KillerWhaleのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 0.0008944293975428763, + "score": 0.0009812203580332143, "project_name": "ArrowPro-7B-KillerWhale-gguf" }, { - "url": "https://huggingface.co/elyza/ELYZA-japanese-Llama-2-7b-fast-instruct", - "downloads": 6434, - "description": "ELYZA-japanese-Llama-2-7bModel DescriptionELYZA-japanese-Llama-2-7b", + "url": "https://huggingface.co/rinna/japanese-gpt2-small", + "downloads": 6555, + "description": "japanese-gpt2-smallThis repository provides a small-sized Japanese GPT-2 model.", "source": "Hugging Face", - "score": 0.0008941514518009425, - "project_name": "ELYZA-japanese-Llama-2-7b-fast-instruct" + "score": 0.0009807714923616528, + "project_name": "japanese-gpt2-small" }, { - "url": "https://huggingface.co/reazon-research/reazonspeech-nemo-v2", - "downloads": 6350, - "description": "reazonspeech-nemo-v2reazonspeech-nemo-v2 is an automatic speech recognition model trainedon ReazonSpeech v2.0 corpus.", + "url": "https://huggingface.co/mmnga/umiyuki-Umievo-itr012-Gleipnir-7B-gguf", + "downloads": 6299, + "description": "umiyuki-Umievo-itr012-Gleipnir-7B-ggufumiyukiさんが公開しているUmievo-itr012-Gleipnir-7Bのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 0.0008824777306397241, - "project_name": "reazonspeech-nemo-v2" + "score": 0.0009424682883884136, + "project_name": "umiyuki-Umievo-itr012-Gleipnir-7B-gguf" }, { - "url": "https://huggingface.co/mmnga/Mistral-7B-Instruct-v0.3-gguf", - "downloads": 6198, - "description": "Mistral-7B-Instruct-v0.3-ggufmistralaiさんが公開しているMistral-7B-Instruct-v0.3のggufフォーマット変換版です。", + "url": "https://huggingface.co/jarvisx17/japanese-sentiment-analysis", + "downloads": 6295, + "description": "japanese-sentiment-analysisThis model was trained from scratch on the chABSA dataset.", "source": "Hugging Face", - "score": 0.0008613538542527574, - "project_name": "Mistral-7B-Instruct-v0.3-gguf" + "score": 0.0009418698008263318, + "project_name": "japanese-sentiment-analysis" }, { - "url": "https://huggingface.co/mmnga/umiyuki-Umievo-itr012-Gleipnir-7B-gguf", - "downloads": 6170, - "description": "umiyuki-Umievo-itr012-Gleipnir-7B-ggufumiyukiさんが公開しているUmievo-itr012-Gleipnir-7Bのggufフォーマット変換版です。", + "url": "https://huggingface.co/rinna/japanese-roberta-base", + "downloads": 6132, + "description": "japanese-roberta-baseThis repository provides a base-sized Japanese RoBERTa model.", "source": "Hugging Face", - "score": 0.0008574626138656847, - "project_name": "umiyuki-Umievo-itr012-Gleipnir-7B-gguf" + "score": 0.0009174814326714959, + "project_name": "japanese-roberta-base" }, { - "url": "https://huggingface.co/ku-nlp/deberta-v2-large-japanese-char-wwm", - "downloads": 6111, - "description": "Model Card for Japanese character-level DeBERTa V2 largeModel descriptionThis is a Japanese DeBERTa V2 large model pre-trained on Japanese Wikipedia, the Japanese portion of CC-100, and the Japanese portion of OSCAR.This model is trained with character-level tokenization and whole word masking.", + "url": "https://huggingface.co/elyza/ELYZA-japanese-Llama-2-7b", + "downloads": 6056, + "description": "ELYZA-japanese-Llama-2-7bModel DescriptionELYZA-japanese-Llama-2-7b", "source": "Hugging Face", - "score": 0.0008492632144786384, - "project_name": "deberta-v2-large-japanese-char-wwm" + "score": 0.0009061101689919405, + "project_name": "ELYZA-japanese-Llama-2-7b" }, { - "url": "https://huggingface.co/elyza/ELYZA-japanese-Llama-2-7b", - "downloads": 6087, + "url": "https://huggingface.co/elyza/ELYZA-japanese-Llama-2-7b-fast-instruct", + "downloads": 5745, "description": "ELYZA-japanese-Llama-2-7bModel DescriptionELYZA-japanese-Llama-2-7b", "source": "Hugging Face", - "score": 0.0008459278655754332, - "project_name": "ELYZA-japanese-Llama-2-7b" + "score": 0.0008595777610400756, + "project_name": "ELYZA-japanese-Llama-2-7b-fast-instruct" + }, + { + "url": "https://huggingface.co/mmnga/ELYZA-japanese-Llama-2-7b-fast-instruct-gguf", + "downloads": 5635, + "description": "ELYZA-japanese-Llama-2-7b-fast-instruct-ggufELYZAさんが公開しているELYZA-japanese-Llama-2-7b-fast-instructのggufフォーマット変換版です。", + "source": "Hugging Face", + "score": 0.0008431193530828244, + "project_name": "ELYZA-japanese-Llama-2-7b-fast-instruct-gguf" + }, + { + "url": "https://huggingface.co/ku-nlp/deberta-v2-large-japanese-char-wwm", + "downloads": 5509, + "description": "Model Card for Japanese character-level DeBERTa V2 largeModel descriptionThis is a Japanese DeBERTa V2 large model pre-trained on Japanese Wikipedia, the Japanese portion of CC-100, and the Japanese portion of OSCAR.This model is trained with character-level tokenization and whole word masking.", + "source": "Hugging Face", + "score": 0.0008242669948772457, + "project_name": "deberta-v2-large-japanese-char-wwm" }, { "url": "https://huggingface.co/elyza/ELYZA-japanese-Llama-2-13b-instruct", - "downloads": 6050, + "downloads": 5432, "description": "ELYZA-japanese-Llama-2-13bModel DescriptionELYZA-japanese-Llama-2-13b は、 Llama 2をベースとして日本語能力を拡張するために追加事前学習を行ったモデルです。", "source": "Hugging Face", - "score": 0.0008407858693496584, + "score": 0.0008127461093071698, "project_name": "ELYZA-japanese-Llama-2-13b-instruct" }, { - "url": "https://huggingface.co/line-corporation/line-distilbert-base-japanese", - "downloads": 5945, - "description": "LINE DistilBERT", + "url": "https://huggingface.co/tokyotech-llm/Swallow-7b-instruct-hf", + "downloads": 5429, + "description": "SwallowOur Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data.", "source": "Hugging Face", - "score": 0.0008261937178981354, - "project_name": "line-distilbert-base-japanese" + "score": 0.0008122972436356085, + "project_name": "Swallow-7b-instruct-hf" }, { "url": "https://huggingface.co/rinna/japanese-gpt2-xsmall", - "downloads": 5611, + "downloads": 5221, "description": "japanese-gpt2-xsmallThis repository provides an extra-small-sized Japanese GPT-2 model.", "source": "Hugging Face", - "score": 0.0007797767789951955, + "score": 0.0007811758904073515, "project_name": "japanese-gpt2-xsmall" }, { - "url": "https://huggingface.co/tokyotech-llm/Swallow-7b-instruct-hf", - "downloads": 5481, - "description": "SwallowOur Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data.", + "url": "https://huggingface.co/reazon-research/reazonspeech-nemo-v2", + "downloads": 5191, + "description": "reazonspeech-nemo-v2reazonspeech-nemo-v2 is an automatic speech recognition model trainedon ReazonSpeech v2.0 corpus.", "source": "Hugging Face", - "score": 0.0007617103057695004, - "project_name": "Swallow-7b-instruct-hf" + "score": 0.0007766872336917376, + "project_name": "reazonspeech-nemo-v2" }, { "url": "https://huggingface.co/FINGU-AI/FinguAI-Chat-v1", - "downloads": 5122, + "downloads": 4830, "description": "FINGU-AI/FinguAI-Chat-v1OverviewThe FINGU-AI/FinguAI-Chat-v1 model offers a specialized curriculum tailored to English, Korean, and Japanese speakers interested in finance, investment, and legal frameworks.", "source": "Hugging Face", - "score": 0.0007118190450923885, + "score": 0.0007226737312138495, "project_name": "FinguAI-Chat-v1" }, { "url": "https://huggingface.co/kotoba-tech/kotoba-whisper-v1.1", - "downloads": 4929, + "downloads": 4730, "description": "Kotoba-Whisper-v1.1Kotoba-Whisper-v1.1 is a Japanese ASR model based on kotoba-tech/kotoba-whisper-v1.0, withadditional postprocessing stacks integrated as pipeline.", "source": "Hugging Face", - "score": 0.0006849972809957795, + "score": 0.000707711542161803, "project_name": "kotoba-whisper-v1.1" }, { - "url": "https://huggingface.co/rinna/japanese-gpt2-small", - "downloads": 4871, - "description": "japanese-gpt2-smallThis repository provides a small-sized Japanese GPT-2 model.", + "url": "https://huggingface.co/sbintuitions/tiny-lm", + "downloads": 4688, + "description": "tiny-lmThis repository provides a tiny 16M parameters language model for debugging and testing purposes.", "source": "Hugging Face", - "score": 0.0006769368544797002, - "project_name": "japanese-gpt2-small" + "score": 0.0007014274227599434, + "project_name": "tiny-lm" }, { - "url": "https://huggingface.co/elyza/ELYZA-japanese-Llama-2-13b", - "downloads": 4764, - "description": "ELYZA-japanese-Llama-2-13bModel DescriptionELYZA-japanese-Llama-2-13b は、 Llama 2をベースとして日本語能力を拡張するために追加事前学習を行ったモデルです。", + "url": "https://huggingface.co/cyberagent/open-calm-small", + "downloads": 4345, + "description": "OpenCALM-SmallModel DescriptionOpenCALM is a suite of decoder-only language models pre-trained on Japanese datasets, developed by", "source": "Hugging Face", - "score": 0.0006620667572862434, - "project_name": "ELYZA-japanese-Llama-2-13b" + "score": 0.0006501071143114236, + "project_name": "open-calm-small" }, { - "url": "https://huggingface.co/tokyotech-llm/Swallow-70b-instruct-hf", - "downloads": 4731, - "description": "SwallowOur Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data.", + "url": "https://huggingface.co/line-corporation/clip-japanese-base", + "downloads": 4287, + "description": "clip-japanese-baseThis is a Japanese CLIP (Contrastive Language-Image Pre-training) model developed by LY Corporation.", "source": "Hugging Face", - "score": 0.0006574806525443362, - "project_name": "Swallow-70b-instruct-hf" + "score": 0.0006414290446612366, + "project_name": "clip-japanese-base" }, { - "url": "https://huggingface.co/jarvisx17/japanese-sentiment-analysis", - "downloads": 4584, - "description": "japanese-sentiment-analysisThis model was trained from scratch on the chABSA dataset.", + "url": "https://huggingface.co/den2nova/FlexDreamHK", + "downloads": 4272, + "description": "🎈 FlexDreamHKFlexDreamHKはリークされたNovelAIモデルの入っていない、あるいはそのリスクを可能な限り低くしたモデルを目指して作成しました。", "source": "Hugging Face", - "score": 0.000637051640512204, - "project_name": "japanese-sentiment-analysis" + "score": 0.0006391847163034297, + "project_name": "FlexDreamHK" }, { - "url": "https://huggingface.co/bclavie/JaColBERT", - "downloads": 4424, - "description": "このドキュメントの日本語版はまだ作成中です。", + "url": "https://huggingface.co/Helsinki-NLP/opus-tatoeba-en-ja", + "downloads": 4065, + "description": "en-jasource group: Englishtarget group: JapaneseOPUS readme: eng-jpnmodel: transformer-alignsource language(s): engtarget language(s): jpnmodel: transformer-alignpre-processing: normalization + SentencePiece (spm32k,spm32k)", "source": "Hugging Face", - "score": 0.0006148159811575023, - "project_name": "JaColBERT" + "score": 0.0006082129849656932, + "project_name": "opus-tatoeba-en-ja" }, { - "url": "https://huggingface.co/tokyotech-llm/Swallow-7b-instruct-v0.1", - "downloads": 4372, - "description": "SwallowOur Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data.", + "url": "https://huggingface.co/elyza/ELYZA-japanese-Llama-2-13b", + "downloads": 3942, + "description": "ELYZA-japanese-Llama-2-13bModel DescriptionELYZA-japanese-Llama-2-13b は、 Llama 2をベースとして日本語能力を拡張するために追加事前学習を行ったモデルです。", "source": "Hugging Face", - "score": 0.0006075893918672242, - "project_name": "Swallow-7b-instruct-v0.1" + "score": 0.0005898094924316759, + "project_name": "ELYZA-japanese-Llama-2-13b" }, { - "url": "https://huggingface.co/den2nova/FlexDreamHK", - "downloads": 4275, - "description": "🎈 FlexDreamHKFlexDreamHKはリークされたNovelAIモデルの入っていない、あるいはそのリスクを可能な限り低くしたモデルを目指して作成しました。", + "url": "https://huggingface.co/tokyotech-llm/Swallow-70b-instruct-hf", + "downloads": 3881, + "description": "SwallowOur Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data.", "source": "Hugging Face", - "score": 0.0005941090233834362, - "project_name": "FlexDreamHK" - }, - { - "url": "https://huggingface.co/tokyotech-llm/Swallow-7b-plus-hf", - "downloads": 4261, - "description": "SwallowOur Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data.", - "source": "Hugging Face", - "score": 0.0005921634031898999, - "project_name": "Swallow-7b-plus-hf" - }, - { - "url": "https://huggingface.co/cyberagent/open-calm-small", - "downloads": 4261, - "description": "OpenCALM-SmallModel DescriptionOpenCALM is a suite of decoder-only language models pre-trained on Japanese datasets, developed by", - "source": "Hugging Face", - "score": 0.0005921634031898999, - "project_name": "open-calm-small" + "score": 0.0005806825571099275, + "project_name": "Swallow-70b-instruct-hf" }, { "url": "https://huggingface.co/cyberagent/open-calm-7b", - "downloads": 4228, + "downloads": 3787, "description": "OpenCALM-7BModel DescriptionOpenCALM is a suite of decoder-only language models pre-trained on Japanese datasets, developed by CyberAgent, Inc.", "source": "Hugging Face", - "score": 0.0005875772984479927, + "score": 0.0005666180994010038, "project_name": "open-calm-7b" }, { - "url": "https://huggingface.co/tokyotech-llm/Swallow-7b-hf", - "downloads": 4213, + "url": "https://huggingface.co/christian-phu/bert-finetuned-japanese-sentiment", + "downloads": 3697, + "description": "bert-finetuned-japanese-sentimentThis model is a fine-tuned version of cl-tohoku/bert-base-japanese-v2 on product amazon reviews japanese dataset.", + "source": "Hugging Face", + "score": 0.0005531521292541618, + "project_name": "bert-finetuned-japanese-sentiment" + }, + { + "url": "https://huggingface.co/tokyotech-llm/Swallow-70b-hf", + "downloads": 3647, "description": "SwallowOur Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data.", "source": "Hugging Face", - "score": 0.0005854927053834894, - "project_name": "Swallow-7b-hf" + "score": 0.0005456710347281385, + "project_name": "Swallow-70b-hf" }, { - "url": "https://huggingface.co/elyza/ELYZA-japanese-Llama-2-13b-fast-instruct", - "downloads": 4059, - "description": "ELYZA-japanese-Llama-2-13b-fast-instructModel DescriptionELYZA-japanese-Llama-2-13b は、 Llama 2をベースとして日本語能力を拡張するために追加事前学習を行ったモデルです。", + "url": "https://huggingface.co/mmnga/Llama-3-70B-japanese-suzume-vector-v0.1", + "downloads": 3640, + "description": "Model Card for Model ID実験モデルです /", "source": "Hugging Face", - "score": 0.000564090883254589, - "project_name": "ELYZA-japanese-Llama-2-13b-fast-instruct" + "score": 0.0005446236814944952, + "project_name": "Llama-3-70B-japanese-suzume-vector-v0.1" }, { "url": "https://huggingface.co/mmnga/Ninja-v1-NSFW-128k-gguf", - "downloads": 3983, + "downloads": 3568, "description": "Ninja-v1-NSFW-128k-ggufLocal-Novel-LLM-projectさんが公開しているNinja-v1-NSFW-128kのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 0.0005535289450611057, + "score": 0.0005338509053770217, "project_name": "Ninja-v1-NSFW-128k-gguf" }, { - "url": "https://huggingface.co/tokyotech-llm/Swallow-70b-hf", - "downloads": 3807, + "url": "https://huggingface.co/tokyotech-llm/Swallow-7b-hf", + "downloads": 3559, "description": "SwallowOur Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data.", "source": "Hugging Face", - "score": 0.0005290697197709338, - "project_name": "Swallow-70b-hf" - }, - { - "url": "https://huggingface.co/Helsinki-NLP/opus-tatoeba-en-ja", - "downloads": 3790, - "description": "en-jasource group: Englishtarget group: JapaneseOPUS readme: eng-jpnmodel: transformer-alignsource language(s): engtarget language(s): jpnmodel: transformer-alignpre-processing: normalization + SentencePiece (spm32k,spm32k)", - "source": "Hugging Face", - "score": 0.0005267071809644967, - "project_name": "opus-tatoeba-en-ja" + "score": 0.0005325043083623376, + "project_name": "Swallow-7b-hf" }, { - "url": "https://huggingface.co/mmnga/Llama-3-70B-japanese-suzume-vector-v0.1", - "downloads": 3578, - "description": "Model Card for Model ID実験モデルです /", + "url": "https://huggingface.co/tohoku-nlp/bert-large-japanese", + "downloads": 3462, + "description": "BERT large Japanese (unidic-lite with whole word masking, jawiki-20200831)This is a BERT model pretrained on texts in the Japanese language.", "source": "Hugging Face", - "score": 0.0004972449323195169, - "project_name": "Llama-3-70B-japanese-suzume-vector-v0.1" + "score": 0.0005179909849818523, + "project_name": "bert-large-japanese" }, { - "url": "https://huggingface.co/megagonlabs/transformers-ud-japanese-electra-base-ginza-510", - "downloads": 3521, - "description": "transformers-ud-japanese-electra-ginza-510 (sudachitra-wordpiece, mC4 Japanese)This is an ELECTRA model pretrained on approximately 200M Japanese sentences extracted from the mC4 and finetuned by spaCy v3 on UD_Japanese_BCCWJ r2.8.The base pretrain model is megagonlabs/transformers-ud-japanese-electra-base-discrimininator.", + "url": "https://huggingface.co/tokyotech-llm/Swallow-7b-plus-hf", + "downloads": 3460, + "description": "SwallowOur Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data.", "source": "Hugging Face", - "score": 0.0004893234786744045, - "project_name": "transformers-ud-japanese-electra-base-ginza-510" + "score": 0.0005176917412008115, + "project_name": "Swallow-7b-plus-hf" }, { - "url": "https://huggingface.co/tohoku-nlp/bert-large-japanese", - "downloads": 3520, - "description": "BERT large Japanese (unidic-lite with whole word masking, jawiki-20200831)This is a BERT model pretrained on texts in the Japanese language.", + "url": "https://huggingface.co/elyza/ELYZA-japanese-Llama-2-13b-fast-instruct", + "downloads": 3412, + "description": "ELYZA-japanese-Llama-2-13b-fast-instructModel DescriptionELYZA-japanese-Llama-2-13b は、 Llama 2をベースとして日本語能力を拡張するために追加事前学習を行ったモデルです。", "source": "Hugging Face", - "score": 0.0004891845058034376, - "project_name": "bert-large-japanese" + "score": 0.0005105098904558291, + "project_name": "ELYZA-japanese-Llama-2-13b-fast-instruct" }, { "url": "https://huggingface.co/ushikado/yuyuyui-chatbot", - "downloads": 3372, + "downloads": 3350, "description": "yuyuyui-chatbotThis model is based on rinna/japanese-gpt2-medium and finetuned on Yuyuyui scenario corpus.", "source": "Hugging Face", - "score": 0.00046861652090033853, + "score": 0.0005012333332435602, "project_name": "yuyuyui-chatbot" }, { "url": "https://huggingface.co/rinna/japanese-hubert-base", - "downloads": 3357, + "downloads": 3313, "description": "rinna/japanese-hubert-baseOverviewThis is a Japanese HuBERT Base model trained by rinna Co.", "source": "Hugging Face", - "score": 0.00046653192783583524, + "score": 0.0004956973232943029, "project_name": "japanese-hubert-base" }, { - "url": "https://huggingface.co/mmnga/ELYZA-japanese-Llama-2-13b-fast-instruct-gguf", - "downloads": 3301, - "description": "ELYZA-japanese-Llama-2-13b-fast-instruct-ggufELYZAさんが公開しているELYZA-japanese-Llama-2-13b-fast-instructのggufフォーマット変換版です。", + "url": "https://huggingface.co/megagonlabs/transformers-ud-japanese-electra-base-ginza-510", + "downloads": 3283, + "description": "transformers-ud-japanese-electra-ginza-510 (sudachitra-wordpiece, mC4 Japanese)This is an ELECTRA model pretrained on approximately 200M Japanese sentences extracted from the mC4 and finetuned by spaCy v3 on UD_Japanese_BCCWJ r2.8.The base pretrain model is megagonlabs/transformers-ud-japanese-electra-base-discrimininator.", "source": "Hugging Face", - "score": 0.00045874944706168964, - "project_name": "ELYZA-japanese-Llama-2-13b-fast-instruct-gguf" + "score": 0.000491208666578689, + "project_name": "transformers-ud-japanese-electra-base-ginza-510" }, { - "url": "https://huggingface.co/rinna/youri-7b", - "downloads": 3296, - "description": "rinna/youri-7bOverviewWe conduct continual pre-training of llama2-7b on 40B tokens from a mixture of Japanese and English datasets.", + "url": "https://huggingface.co/mmnga/ELYZA-japanese-Llama-2-7b-instruct-gguf", + "downloads": 3191, + "description": "ELYZA-japanese-Llama-2-7b-instruct-ggufELYZAさんが公開しているELYZA-japanese-Llama-2-7b-instructのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 0.0004580545827068552, - "project_name": "youri-7b" + "score": 0.00047744345265080614, + "project_name": "ELYZA-japanese-Llama-2-7b-instruct-gguf" }, { - "url": "https://huggingface.co/sbintuitions/tiny-lm", - "downloads": 3279, - "description": "tiny-lmThis repository provides a tiny 16M parameters language model for debugging and testing purposes.", + "url": "https://huggingface.co/MCZK/Ninja-V2-7B-GGUF", + "downloads": 3168, + "description": "Local-Novel-LLM-project様の Ninja-V2-7B をGGUF形式に変換したものです。", "source": "Hugging Face", - "score": 0.00045569204390041813, - "project_name": "tiny-lm" + "score": 0.00047400214916883544, + "project_name": "Ninja-V2-7B-GGUF" }, { - "url": "https://huggingface.co/mmnga/ELYZA-japanese-Llama-2-7b-instruct-gguf", - "downloads": 3242, - "description": "ELYZA-japanese-Llama-2-7b-instruct-ggufELYZAさんが公開しているELYZA-japanese-Llama-2-7b-instructのggufフォーマット変換版です。", + "url": "https://huggingface.co/MCZK/Vecteus-V2-7B-GGUF", + "downloads": 3142, + "description": "Local-Novel-LLM-project様の Vecteus-V2-7B をGGUF形式に変換したものです。", "source": "Hugging Face", - "score": 0.0004505500476746434, - "project_name": "ELYZA-japanese-Llama-2-7b-instruct-gguf" + "score": 0.00047011198001530334, + "project_name": "Vecteus-V2-7B-GGUF" }, { - "url": "https://huggingface.co/nlp-waseda/roberta-base-japanese-with-auto-jumanpp", - "downloads": 3170, - "description": "nlp-waseda/roberta-base-japanese-with-auto-jumanppModel descriptionThis is a Japanese RoBERTa base model pretrained on Japanese Wikipedia and the Japanese portion of CC-100.How to useYou can use this model for masked language modeling as follows:from transformers import AutoTokenizer, AutoModelForMaskedLMtokenizer = AutoTokenizer.from_pretrained(\"nlp-waseda/roberta-base-japanese-with-auto-jumanpp\")", + "url": "https://huggingface.co/stockmark/stockmark-13b", + "downloads": 3083, + "description": "stockmark/stockmark-13bStockmark-13b is a 13 billion parameter LLM pretrained from scratch based on Japanese corpus of about 220B tokens.", "source": "Hugging Face", - "score": 0.0004405440009650276, - "project_name": "roberta-base-japanese-with-auto-jumanpp" + "score": 0.00046128428847459584, + "project_name": "stockmark-13b" + }, + { + "url": "https://huggingface.co/mmnga/ELYZA-japanese-Llama-2-13b-fast-instruct-gguf", + "downloads": 3074, + "description": "ELYZA-japanese-Llama-2-13b-fast-instruct-ggufELYZAさんが公開しているELYZA-japanese-Llama-2-13b-fast-instructのggufフォーマット変換版です。", + "source": "Hugging Face", + "score": 0.0004599376914599117, + "project_name": "ELYZA-japanese-Llama-2-13b-fast-instruct-gguf" }, { "url": "https://huggingface.co/Vsukiyaki/Yaki-Dofu-Mix", - "downloads": 3155, + "downloads": 3031, "description": "Yaki-Dofu-Mix概要 / OverviewYaki-Dofu-Mixは、アニメ風の画風に特化したマージモデルです。 ", "source": "Hugging Face", - "score": 0.0004384594079005243, + "score": 0.00045350395016753163, "project_name": "Yaki-Dofu-Mix" }, { - "url": "https://huggingface.co/stockmark/stockmark-13b", - "downloads": 3148, - "description": "stockmark/stockmark-13bStockmark-13b is a 13 billion parameter LLM pretrained from scratch based on Japanese corpus of about 220B tokens.", + "url": "https://huggingface.co/alabnii/jmedroberta-base-sentencepiece-vocab50000", + "downloads": 2909, + "description": "alabnii/jmedroberta-base-sentencepiece-vocab50000Model descriptionThis is a Japanese RoBERTa base model pre-trained on academic articles in medical sciences collected by Japan Science and Technology Agency (JST).", "source": "Hugging Face", - "score": 0.00043748659780375616, - "project_name": "stockmark-13b" + "score": 0.0004352500795240348, + "project_name": "jmedroberta-base-sentencepiece-vocab50000" }, { - "url": "https://huggingface.co/oshizo/sbert-jsnli-luke-japanese-base-lite", - "downloads": 3079, - "description": "sbert-jsnli-luke-japanese-base-liteThis is a sentence-transformers model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search.", + "url": "https://huggingface.co/QuantFactory/Oumuamua-7b-instruct-v2-GGUF", + "downloads": 2869, + "description": "Oumuamua-7b-instruct-v2-GGUFThis is quantized version of nitky/Oumuamua-7b-instruct-v2 created using llama.cppModel Description🚨 If you want to avoid outputs that appear to be literal translations, please prompt this model to role-play as a Japanese person.", "source": "Hugging Face", - "score": 0.000427897469707041, - "project_name": "sbert-jsnli-luke-japanese-base-lite" + "score": 0.00042926520390321616, + "project_name": "Oumuamua-7b-instruct-v2-GGUF" }, { - "url": "https://huggingface.co/MCZK/Vecteus-V2-7B-GGUF", - "downloads": 3072, - "description": "Local-Novel-LLM-project様の Vecteus-V2-7B をGGUF形式に変換したものです。", + "url": "https://huggingface.co/rinna/japanese-gpt-neox-3.6b-instruction-sft", + "downloads": 2825, + "description": "japanese-gpt-neox-3.6b-instruction-sftOverviewThis repository provides a Japanese GPT-NeoX model of 3.6 billion parameters.", "source": "Hugging Face", - "score": 0.00042692465961027284, - "project_name": "Vecteus-V2-7B-GGUF" + "score": 0.0004226818407203157, + "project_name": "japanese-gpt-neox-3.6b-instruction-sft" }, { - "url": "https://huggingface.co/mmnga/ELYZA-japanese-Llama-2-7b-fast-instruct-gguf", - "downloads": 3064, - "description": "ELYZA-japanese-Llama-2-7b-fast-instruct-ggufELYZAさんが公開しているELYZA-japanese-Llama-2-7b-fast-instructのggufフォーマット変換版です。", + "url": "https://huggingface.co/cyberagent/calm2-7b", + "downloads": 2640, + "description": "CyberAgentLM2-7B (CALM2-7B)", "source": "Hugging Face", - "score": 0.0004258128766425377, - "project_name": "ELYZA-japanese-Llama-2-7b-fast-instruct-gguf" + "score": 0.0003950017909740295, + "project_name": "calm2-7b" }, { - "url": "https://huggingface.co/MCZK/Ninja-V2-7B-GGUF", - "downloads": 3026, - "description": "Local-Novel-LLM-project様の Ninja-V2-7B をGGUF形式に変換したものです。", + "url": "https://huggingface.co/rinna/youri-7b", + "downloads": 2637, + "description": "rinna/youri-7bOverviewWe conduct continual pre-training of llama2-7b on 40B tokens from a mixture of Japanese and English datasets.", "source": "Hugging Face", - "score": 0.0004205319075457961, - "project_name": "Ninja-V2-7B-GGUF" + "score": 0.0003945529253024681, + "project_name": "youri-7b" }, { "url": "https://huggingface.co/stabilityai/japanese-stablelm-instruct-gamma-7b", - "downloads": 3024, + "downloads": 2624, "description": "Japanese Stable LM Instruct Gamma 7BModel", "source": "Hugging Face", - "score": 0.0004202539618038623, + "score": 0.00039260784072570206, "project_name": "japanese-stablelm-instruct-gamma-7b" }, { - "url": "https://huggingface.co/line-corporation/clip-japanese-base", - "downloads": 2992, - "description": "clip-japanese-baseThis is a Japanese CLIP (Contrastive Language-Image Pre-training) model developed by LY Corporation.", + "url": "https://huggingface.co/rinna/japanese-gpt-1b", + "downloads": 2555, + "description": "japanese-gpt-1bThis repository provides a 1.3B-parameter Japanese GPT model.", "source": "Hugging Face", - "score": 0.000415806829932922, - "project_name": "clip-japanese-base" + "score": 0.00038228393027978997, + "project_name": "japanese-gpt-1b" }, { - "url": "https://huggingface.co/alabnii/jmedroberta-base-sentencepiece-vocab50000", - "downloads": 2978, - "description": "alabnii/jmedroberta-base-sentencepiece-vocab50000Model descriptionThis is a Japanese RoBERTa base model pre-trained on academic articles in medical sciences collected by Japan Science and Technology Agency (JST).", + "url": "https://huggingface.co/cyberagent/open-calm-large", + "downloads": 2503, + "description": "OpenCALM-LargeModel DescriptionOpenCALM is a suite of decoder-only language models pre-trained on Japanese datasets, developed by", "source": "Hugging Face", - "score": 0.00041386120973938554, - "project_name": "jmedroberta-base-sentencepiece-vocab50000" + "score": 0.0003745035919727257, + "project_name": "open-calm-large" }, { "url": "https://huggingface.co/mmnga/aibuncho-japanese-novel-gpt-j-6b-gguf", - "downloads": 2934, + "downloads": 2479, "description": "AIBunCho/japanese-novel-gpt-j-6bAI BunChoさんが公開しているjapanese-novel-gpt-j-6bのgguf変換版です。", "source": "Hugging Face", - "score": 0.0004077464034168426, + "score": 0.0003709126666002345, "project_name": "aibuncho-japanese-novel-gpt-j-6b-gguf" }, { - "url": "https://huggingface.co/rinna/bilingual-gpt-neox-4b", - "downloads": 2879, - "description": "bilingual-gpt-neox-4bOverviewThis repository provides an English-Japanese bilingual GPT-NeoX model of 3.8 billion parameters.", + "url": "https://huggingface.co/oshizo/sbert-jsnli-luke-japanese-base-lite", + "downloads": 2430, + "description": "sbert-jsnli-luke-japanese-base-liteThis is a sentence-transformers model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search.", "source": "Hugging Face", - "score": 0.0004001028955136639, - "project_name": "bilingual-gpt-neox-4b" + "score": 0.00036358119396473173, + "project_name": "sbert-jsnli-luke-japanese-base-lite" }, { - "url": "https://huggingface.co/cyberagent/open-calm-large", - "downloads": 2875, - "description": "OpenCALM-LargeModel DescriptionOpenCALM is a suite of decoder-only language models pre-trained on Japanese datasets, developed by", + "url": "https://huggingface.co/rinna/japanese-cloob-vit-b-16", + "downloads": 2369, + "description": "rinna/japanese-cloob-vit-b-16This is a Japanese CLOOB (Contrastive Leave One Out Boost) model trained by rinna Co.", "source": "Hugging Face", - "score": 0.00039954700402979634, - "project_name": "open-calm-large" + "score": 0.0003544542586429833, + "project_name": "japanese-cloob-vit-b-16" }, { - "url": "https://huggingface.co/rinna/japanese-gpt-neox-3.6b-instruction-sft", - "downloads": 2775, - "description": "japanese-gpt-neox-3.6b-instruction-sftOverviewThis repository provides a Japanese GPT-NeoX model of 3.6 billion parameters.", + "url": "https://huggingface.co/rinna/japanese-gpt-neox-3.6b-instruction-ppo", + "downloads": 2348, + "description": "japanese-gpt-neox-3.6b-instruction-ppoOverviewThis repository provides a Japanese GPT-NeoX model of 3.6 billion parameters.", "source": "Hugging Face", - "score": 0.0003856497169331078, - "project_name": "japanese-gpt-neox-3.6b-instruction-sft" + "score": 0.00035131219894205353, + "project_name": "japanese-gpt-neox-3.6b-instruction-ppo" }, { - "url": "https://huggingface.co/rinna/japanese-gpt-1b", - "downloads": 2774, - "description": "japanese-gpt-1bThis repository provides a 1.3B-parameter Japanese GPT model.", + "url": "https://huggingface.co/nlp-waseda/roberta-base-japanese-with-auto-jumanpp", + "downloads": 2337, + "description": "nlp-waseda/roberta-base-japanese-with-auto-jumanppModel descriptionThis is a Japanese RoBERTa base model pretrained on Japanese Wikipedia and the Japanese portion of CC-100.How to useYou can use this model for masked language modeling as follows:from transformers import AutoTokenizer, AutoModelForMaskedLMtokenizer = AutoTokenizer.from_pretrained(\"nlp-waseda/roberta-base-japanese-with-auto-jumanpp\")", "source": "Hugging Face", - "score": 0.0003855107440621409, - "project_name": "japanese-gpt-1b" + "score": 0.0003496663581463284, + "project_name": "roberta-base-japanese-with-auto-jumanpp" }, { - "url": "https://huggingface.co/QuantFactory/Oumuamua-7b-instruct-v2-GGUF", - "downloads": 2737, - "description": "Oumuamua-7b-instruct-v2-GGUFThis is quantized version of nitky/Oumuamua-7b-instruct-v2 created using llama.cppModel Description🚨 If you want to avoid outputs that appear to be literal translations, please prompt this model to role-play as a Japanese person.", + "url": "https://huggingface.co/QuantFactory/shisa-gamma-7b-v1-GGUF", + "downloads": 2311, + "description": "QuantFactory/shisa-gamma-7b-v1-GGUFThis is quantized version of augmxnt/shisa-gamma-7b-v1 created using llama.cppModel DescriptionFor more information see our main Shisa 7B modelWe applied a version of our fine-tune data set onto Japanese Stable LM Base Gamma 7B and it performed pretty well, just sharing since it might be of interest.", "source": "Hugging Face", - "score": 0.0003803687478363661, - "project_name": "Oumuamua-7b-instruct-v2-GGUF" + "score": 0.0003457761889927963, + "project_name": "shisa-gamma-7b-v1-GGUF" }, { - "url": "https://huggingface.co/elyza/ELYZA-japanese-Llama-2-13b-fast", - "downloads": 2656, - "description": "ELYZA-japanese-Llama-2-13b-fastModel DescriptionELYZA-japanese-Llama-2-13b は、 Llama 2をベースとして日本語能力を拡張するために追加事前学習を行ったモデルです。", + "url": "https://huggingface.co/QuantFactory/Umievo-itr012-Gleipnir-7B-GGUF", + "downloads": 2296, + "description": "Umievo-itr012-Gleipnir-7B-GGUFThis is quantized version of umiyuki/Umievo-itr012-Gleipnir-7B created using llama.cppModel Descriptionこのモデルは強力な４つの日本語モデルを進化的アルゴリズムで進化的マージしたものです。", "source": "Hugging Face", - "score": 0.0003691119452880484, - "project_name": "ELYZA-japanese-Llama-2-13b-fast" + "score": 0.00034353186063498933, + "project_name": "Umievo-itr012-Gleipnir-7B-GGUF" }, { - "url": "https://huggingface.co/rinna/japanese-gpt-neox-3.6b-instruction-ppo", - "downloads": 2607, - "description": "japanese-gpt-neox-3.6b-instruction-ppoOverviewThis repository provides a Japanese GPT-NeoX model of 3.6 billion parameters.", + "url": "https://huggingface.co/rinna/bilingual-gpt-neox-4b", + "downloads": 2290, + "description": "bilingual-gpt-neox-4bOverviewThis repository provides an English-Japanese bilingual GPT-NeoX model of 3.8 billion parameters.", "source": "Hugging Face", - "score": 0.00036230227461067096, - "project_name": "japanese-gpt-neox-3.6b-instruction-ppo" + "score": 0.0003426341292918665, + "project_name": "bilingual-gpt-neox-4b" + }, + { + "url": "https://huggingface.co/QuantFactory/Oumuamua-7b-instruct-GGUF", + "downloads": 2287, + "description": "Oumuamua-7b-instruct-GGUFThis is quantized version of nitky/Oumuamua-7b-instruct created using llama.cppModel DescriptionThis is a merge of pre-trained language models created using mergekit.Output example[INST] <<SYS>>あなたは日本語を話す優秀なアシスタントです。", + "source": "Hugging Face", + "score": 0.0003421852636203051, + "project_name": "Oumuamua-7b-instruct-GGUF" }, { "url": "https://huggingface.co/pkshatech/simcse-ja-bert-base-clcmlp", - "downloads": 2565, + "downloads": 2213, "description": "Japanese SimCSE (BERT-base)", "source": "Hugging Face", - "score": 0.0003564654140300618, + "score": 0.00033111324372179065, "project_name": "simcse-ja-bert-base-clcmlp" }, { - "url": "https://huggingface.co/rinna/japanese-cloob-vit-b-16", - "downloads": 2520, - "description": "rinna/japanese-cloob-vit-b-16This is a Japanese CLOOB (Contrastive Leave One Out Boost) model trained by rinna Co.", + "url": "https://huggingface.co/stabilityai/japanese-stablelm-instruct-beta-70b", + "downloads": 2195, + "description": "Japanese-StableLM-Instruct-Beta-70BA cute robot wearing a kimono writes calligraphy with one single brush — Stable Diffusion XLModel Descriptionjapanese-stablelm-instruct-beta-70b is a 70B-parameter decoder-only language model based on japanese-stablelm-base-beta-70b and further fine tuned on Databricks Dolly-15k, Anthropic HH, and other public data.", "source": "Hugging Face", - "score": 0.00035021163483655194, - "project_name": "japanese-cloob-vit-b-16" + "score": 0.00032842004969242227, + "project_name": "japanese-stablelm-instruct-beta-70b" }, { "url": "https://huggingface.co/rinna/bilingual-gpt-neox-4b-8k", - "downloads": 2485, + "downloads": 2180, "description": "bilingual-gpt-neox-4b-8kOverviewNotice: This model requires transformers>=4.31.0 to work properly.", "source": "Hugging Face", - "score": 0.00034534758435271094, + "score": 0.0003261757213346153, "project_name": "bilingual-gpt-neox-4b-8k" }, { - "url": "https://huggingface.co/Aratako/Ninja-v1-RP-expressive-GGUF", - "downloads": 2453, - "description": "Ninja-v1-RP-expressive-GGUF概要Aratako/Ninja-v1-RP-expressiveの量子化済みGGUF版です。", - "source": "Hugging Face", - "score": 0.0003409004524817706, - "project_name": "Ninja-v1-RP-expressive-GGUF" - }, - { - "url": "https://huggingface.co/mmnga/Vecteus-v1-gguf", - "downloads": 2363, - "description": "Vecteus-v1-ggufLocal-Novel-LLM-projectさんが公開しているVecteus-v1のggufフォーマット変換版です。", - "source": "Hugging Face", - "score": 0.00032839289409475087, - "project_name": "Vecteus-v1-gguf" - }, - { - "url": "https://huggingface.co/QuantFactory/shisa-gamma-7b-v1-GGUF", - "downloads": 2289, - "description": "QuantFactory/shisa-gamma-7b-v1-GGUFThis is quantized version of augmxnt/shisa-gamma-7b-v1 created using llama.cppModel DescriptionFor more information see our main Shisa 7B modelWe applied a version of our fine-tune data set onto Japanese Stable LM Base Gamma 7B and it performed pretty well, just sharing since it might be of interest.", + "url": "https://huggingface.co/ku-nlp/deberta-v2-tiny-japanese-char-wwm", + "downloads": 2123, + "description": "Model Card for Japanese character-level DeBERTa V2 tinyModel descriptionThis is a Japanese DeBERTa V2 tiny model pre-trained on Japanese Wikipedia, the Japanese portion of CC-100, and the Japanese portion of OSCAR.This model is trained with character-level tokenization and whole word masking.", "source": "Hugging Face", - "score": 0.00031810890164320134, - "project_name": "shisa-gamma-7b-v1-GGUF" + "score": 0.0003176472735749487, + "project_name": "deberta-v2-tiny-japanese-char-wwm" }, { - "url": "https://huggingface.co/tohoku-nlp/bert-base-japanese-char-v3", - "downloads": 2278, - "description": "BERT base Japanese (character-level tokenization with whole word masking, CC-100 and jawiki-20230102)This is a BERT model pretrained on texts in the Japanese language.", + "url": "https://huggingface.co/elyza/ELYZA-japanese-Llama-2-13b-fast", + "downloads": 2010, + "description": "ELYZA-japanese-Llama-2-13b-fastModel DescriptionELYZA-japanese-Llama-2-13b は、 Llama 2をベースとして日本語能力を拡張するために追加事前学習を行ったモデルです。", "source": "Hugging Face", - "score": 0.0003165802000625656, - "project_name": "bert-base-japanese-char-v3" + "score": 0.0003007399999461361, + "project_name": "ELYZA-japanese-Llama-2-13b-fast" }, { - "url": "https://huggingface.co/augmxnt/shisa-base-7b-v1", - "downloads": 2266, - "description": "shisa-base-7b-v1shisa-base-7b-v1 takes Mistral 7B and adds an additional 8B tokens of primarily Japanese pre-training.", + "url": "https://huggingface.co/mmnga/Ninja-v1-NSFW-gguf", + "downloads": 1961, + "description": "Ninja-v1-NSFW-ggufLocal-Novel-LLM-projectさんが公開しているNinja-v1-NSFWのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 0.00031491252561096294, - "project_name": "shisa-base-7b-v1" + "score": 0.0002934085273106333, + "project_name": "Ninja-v1-NSFW-gguf" }, { "url": "https://huggingface.co/augmxnt/shisa-7b-v1", - "downloads": 2244, + "downloads": 1893, "description": "Shisa 7BShisa 7B (shisa-7b-v1)", "source": "Hugging Face", - "score": 0.0003118551224496915, + "score": 0.00028323423875524164, "project_name": "shisa-7b-v1" }, { - "url": "https://huggingface.co/QuantFactory/Umievo-itr012-Gleipnir-7B-GGUF", - "downloads": 2232, - "description": "Umievo-itr012-Gleipnir-7B-GGUFThis is quantized version of umiyuki/Umievo-itr012-Gleipnir-7B created using llama.cppModel Descriptionこのモデルは強力な４つの日本語モデルを進化的アルゴリズムで進化的マージしたものです。", + "url": "https://huggingface.co/mm/japanese-e5-mistral-7b_slerp_gguf", + "downloads": 1861, + "description": "Japanese E5 Mixtral 7B Slerp GGUFGGUF conversion of oshizo/japanese-e5-mistral-7b_slerpAvaiable formats:Q2_K.ggufQ3_K.ggufQ4_K.ggufQ5_K.ggufQ6_K.ggufQ8_0.ggufF16.ggufUsageRequires: llama-cpp-pythonfrom functools import partialimport numpy as npfrom llama_cpp import Llamamax_length = 512model = Llama.from_pretrained(repo_id=\"mm/japanese-e5-mistral-7b_slerp_gguf\",filename=\"*Q4_K.gguf\", # Choose from the avaiable formats,embedding=True,n_ctx=max_length,n_batch=max_length,verbose=False,)model.tokenize = partia", "source": "Hugging Face", - "score": 0.00031018744799808884, - "project_name": "Umievo-itr012-Gleipnir-7B-GGUF" + "score": 0.00027844633825858673, + "project_name": "japanese-e5-mistral-7b_slerp_gguf" }, { - "url": "https://huggingface.co/haqishen/Llama-3-8B-Japanese-Instruct", - "downloads": 2182, - "description": "IntroductionWho am I: Qishen Ha", + "url": "https://huggingface.co/rinna/bilingual-gpt-neox-4b-instruction-ppo", + "downloads": 1861, + "description": "bilingual-gpt-neox-4b-instruction-ppoOverviewThis repository provides an English-Japanese bilingual GPT-NeoX model of 3.8 billion parameters.", "source": "Hugging Face", - "score": 0.00030323880444974456, - "project_name": "Llama-3-8B-Japanese-Instruct" + "score": 0.00027844633825858673, + "project_name": "bilingual-gpt-neox-4b-instruction-ppo" }, { - "url": "https://huggingface.co/MCZK/ArrowPro-7B-RobinHood-GGUF", - "downloads": 2174, - "description": "DataPilot様の ArrowPro-7B-RobinHood をGGUF形式に変換したものです。", + "url": "https://huggingface.co/augmxnt/shisa-base-7b-v1", + "downloads": 1847, + "description": "shisa-base-7b-v1shisa-base-7b-v1 takes Mistral 7B and adds an additional 8B tokens of primarily Japanese pre-training.", "source": "Hugging Face", - "score": 0.0003021270214820095, - "project_name": "ArrowPro-7B-RobinHood-GGUF" + "score": 0.0002763516317913002, + "project_name": "shisa-base-7b-v1" }, { - "url": "https://huggingface.co/QuantFactory/Oumuamua-7b-instruct-GGUF", - "downloads": 2170, - "description": "Oumuamua-7b-instruct-GGUFThis is quantized version of nitky/Oumuamua-7b-instruct created using llama.cppModel DescriptionThis is a merge of pre-trained language models created using mergekit.Output example[INST] <<SYS>>あなたは日本語を話す優秀なアシスタントです。", + "url": "https://huggingface.co/mmnga/Vecteus-v1-gguf", + "downloads": 1832, + "description": "Vecteus-v1-ggufLocal-Novel-LLM-projectさんが公開しているVecteus-v1のggufフォーマット変換版です。", "source": "Hugging Face", - "score": 0.0003015711299981419, - "project_name": "Oumuamua-7b-instruct-GGUF" + "score": 0.0002741073034334932, + "project_name": "Vecteus-v1-gguf" }, { - "url": "https://huggingface.co/mmnga/Ninja-v1-NSFW-gguf", - "downloads": 2150, - "description": "Ninja-v1-NSFW-ggufLocal-Novel-LLM-projectさんが公開しているNinja-v1-NSFWのggufフォーマット変換版です。", + "url": "https://huggingface.co/llm-book/bert-base-japanese-v3-unsup-simcse-jawiki", + "downloads": 1829, + "description": "bert-base-japanese-v3-unsup-simcse-jawiki「大規模言語モデル入門」の第8章で紹介している教師なしSimCSEのモデルです。", "source": "Hugging Face", - "score": 0.0002987916725788042, - "project_name": "Ninja-v1-NSFW-gguf" + "score": 0.0002736584377619318, + "project_name": "bert-base-japanese-v3-unsup-simcse-jawiki" }, { - "url": "https://huggingface.co/OrionStarAI/Orion-14B-Chat", - "downloads": 2098, - "description": "Orion-14B🌐English | 🇨🇳中文 | 🇯🇵日本語 | 🇰🇷한국어🤗", + "url": "https://huggingface.co/Mizuiro-sakura/luke-japanese-large-sentiment-analysis-wrime", + "downloads": 1808, + "description": "このモデルはLuke-japanese-large-liteをファインチューニングしたものです。", "source": "Hugging Face", - "score": 0.00029156508328852617, - "project_name": "Orion-14B-Chat" + "score": 0.00027051637806100204, + "project_name": "luke-japanese-large-sentiment-analysis-wrime" }, { - "url": "https://huggingface.co/Lasorco/lametta", - "downloads": 2029, - "description": "このモデルは何？", - "source": "Hugging Face", - "score": 0.0002819759551918111, - "project_name": "lametta" - }, - { - "url": "https://huggingface.co/cyberagent/calm2-7b", - "downloads": 1990, - "description": "CyberAgentLM2-7B (CALM2-7B)", - "source": "Hugging Face", - "score": 0.0002765560132241025, - "project_name": "calm2-7b" - }, - { - "url": "https://huggingface.co/christian-phu/bert-finetuned-japanese-sentiment", - "downloads": 1985, - "description": "bert-finetuned-japanese-sentimentThis model is a fine-tuned version of cl-tohoku/bert-base-japanese-v2 on product amazon reviews japanese dataset.", - "source": "Hugging Face", - "score": 0.00027586114886926806, - "project_name": "bert-finetuned-japanese-sentiment" - }, - { - "url": "https://huggingface.co/stabilityai/japanese-stablelm-instruct-beta-70b", - "downloads": 1971, - "description": "Japanese-StableLM-Instruct-Beta-70BA cute robot wearing a kimono writes calligraphy with one single brush — Stable Diffusion XLModel Descriptionjapanese-stablelm-instruct-beta-70b is a 70B-parameter decoder-only language model based on japanese-stablelm-base-beta-70b and further fine tuned on Databricks Dolly-15k, Anthropic HH, and other public data.", + "url": "https://huggingface.co/nlp-waseda/roberta-base-japanese", + "downloads": 1746, + "description": "nlp-waseda/roberta-base-japaneseModel descriptionThis is a Japanese RoBERTa base model pretrained on Japanese Wikipedia and the Japanese portion of CC-100.How to useYou can use this model for masked language modeling as follows:from transformers import AutoTokenizer, AutoModelForMaskedLMtokenizer = AutoTokenizer.from_pretrained(\"nlp-waseda/roberta-base-japanese\")", "source": "Hugging Face", - "score": 0.0002739155286757317, - "project_name": "japanese-stablelm-instruct-beta-70b" + "score": 0.0002612398208487332, + "project_name": "roberta-base-japanese" }, { - "url": "https://huggingface.co/rinna/bilingual-gpt-neox-4b-instruction-ppo", - "downloads": 1963, - "description": "bilingual-gpt-neox-4b-instruction-ppoOverviewThis repository provides an English-Japanese bilingual GPT-NeoX model of 3.8 billion parameters.", + "url": "https://huggingface.co/cyberagent/open-calm-1b", + "downloads": 1745, + "description": "OpenCALM-1BModel DescriptionOpenCALM is a suite of decoder-only language models pre-trained on Japanese datasets, developed by CyberAgent, Inc.", "source": "Hugging Face", - "score": 0.0002728037457079966, - "project_name": "bilingual-gpt-neox-4b-instruction-ppo" + "score": 0.0002610901989582127, + "project_name": "open-calm-1b" }, { - "url": "https://huggingface.co/Mizuiro-sakura/luke-japanese-large-sentiment-analysis-wrime", - "downloads": 1947, - "description": "このモデルはLuke-japanese-large-liteをファインチューニングしたものです。", + "url": "https://huggingface.co/QuantFactory/shisa-7b-v1-GGUF", + "downloads": 1738, + "description": "QuantFactory/shisa-7b-v1-GGUFThis is quantized version of augmxnt/shisa-base-7b-v1 created using llama.cppModel Descriptionshisa-base-7b-v1 takes Mistral 7B and adds an additional 8B tokens of primarily Japanese pre-training.", "source": "Hugging Face", - "score": 0.00027058017977252643, - "project_name": "luke-japanese-large-sentiment-analysis-wrime" + "score": 0.00026004284572456947, + "project_name": "shisa-7b-v1-GGUF" }, { - "url": "https://huggingface.co/llm-book/bert-base-japanese-v3-unsup-simcse-jawiki", - "downloads": 1837, - "description": "bert-base-japanese-v3-unsup-simcse-jawiki「大規模言語モデル入門」の第8章で紹介している教師なしSimCSEのモデルです。", + "url": "https://huggingface.co/TKU410410103/wav2vec2-base-japanese-asr", + "downloads": 1724, + "description": "wav2vec2-base-asrThis model is a fine-tuned version of rinna/japanese-wav2vec2-base on the common_voice_11_0 dataset for ASR tasks.", "source": "Hugging Face", - "score": 0.000255293163966169, - "project_name": "bert-base-japanese-v3-unsup-simcse-jawiki" + "score": 0.00025794813925728293, + "project_name": "wav2vec2-base-japanese-asr" }, { - "url": "https://huggingface.co/mm/japanese-e5-mistral-7b_slerp_gguf", - "downloads": 1823, - "description": "Japanese E5 Mixtral 7B Slerp GGUFGGUF conversion of oshizo/japanese-e5-mistral-7b_slerpAvaiable formats:Q2_K.ggufQ3_K.ggufQ4_K.ggufQ5_K.ggufQ6_K.ggufQ8_0.ggufF16.ggufUsageRequires: llama-cpp-pythonfrom functools import partialimport numpy as npfrom llama_cpp import Llamamax_length = 512model = Llama.from_pretrained(repo_id=\"mm/japanese-e5-mistral-7b_slerp_gguf\",filename=\"*Q4_K.gguf\", # Choose from the avaiable formats,embedding=True,n_ctx=max_length,n_batch=max_length,verbose=False,)model.tokenize = partia", + "url": "https://huggingface.co/tohoku-nlp/bert-base-japanese-char-v3", + "downloads": 1690, + "description": "BERT base Japanese (character-level tokenization with whole word masking, CC-100 and jawiki-20230102)This is a BERT model pretrained on texts in the Japanese language.", "source": "Hugging Face", - "score": 0.0002533475437726326, - "project_name": "japanese-e5-mistral-7b_slerp_gguf" + "score": 0.0002528609949795871, + "project_name": "bert-base-japanese-char-v3" }, { - "url": "https://huggingface.co/Aratako/Ninja-v1-RP-expressive-v2-GGUF", - "downloads": 1807, - "description": "Ninja-v1-RP-expressive-GGUF概要Aratako/Ninja-v1-RP-expressive-v2の量子化済みGGUF版です。", + "url": "https://huggingface.co/tokyotech-llm/Swallow-13b-instruct-hf", + "downloads": 1642, + "description": "SwallowOur Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data.", "source": "Hugging Face", - "score": 0.00025112397783716244, - "project_name": "Ninja-v1-RP-expressive-v2-GGUF" + "score": 0.00024567914423460473, + "project_name": "Swallow-13b-instruct-hf" }, { - "url": "https://huggingface.co/Fugaku-LLM/Fugaku-LLM-13B-instruct", - "downloads": 1772, - "description": "Fugaku-LLM利用規約この利用規約（以下「本規約」といいます）は、富士通株式会社、国立研究開発法人理化学研究所、国立大学法人東京工業大学、国立大学法人東北大学、株式会社サイバーエージェント、国立大学法人東海国立大学機構、及び株式会社Kotoba Technologies Japan (以下「開発者」といいます)による、スーパーコンピュータ「富岳」政策対応枠における大規模言語モデル分散並列学習手法の開発の成果物として公開する大規模言語モデル（以下「Fugaku-LLM」といいます）の利用に関する条件を定めるものです。", + "url": "https://huggingface.co/line-corporation/japanese-large-lm-3.6b", + "downloads": 1615, + "description": "japanese-large-lm-3.6bThis repository provides a 3.6B parameters Japanese language model, trained by LINE Corporation.", "source": "Hugging Face", - "score": 0.00024625992735332144, - "project_name": "Fugaku-LLM-13B-instruct" + "score": 0.00024163935319055214, + "project_name": "japanese-large-lm-3.6b" }, { - "url": "https://huggingface.co/cyberagent/open-calm-1b", - "downloads": 1724, - "description": "OpenCALM-1BModel DescriptionOpenCALM is a suite of decoder-only language models pre-trained on Japanese datasets, developed by CyberAgent, Inc.", + "url": "https://huggingface.co/line-corporation/japanese-large-lm-1.7b", + "downloads": 1614, + "description": "japanese-large-lm-1.7bThis repository provides a 1.7B parameters Japanese language model, trained by LINE Corporation.", "source": "Hugging Face", - "score": 0.00023958922954691092, - "project_name": "open-calm-1b" + "score": 0.00024148973130003169, + "project_name": "japanese-large-lm-1.7b" }, { - "url": "https://huggingface.co/tokyotech-llm/Swallow-13b-instruct-hf", - "downloads": 1723, - "description": "SwallowOur Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data.", + "url": "https://huggingface.co/Lasorco/lametta", + "downloads": 1600, + "description": "このモデルは何？", "source": "Hugging Face", - "score": 0.00023945025667594403, - "project_name": "Swallow-13b-instruct-hf" + "score": 0.00023939502483274518, + "project_name": "lametta" }, { - "url": "https://huggingface.co/OrionStarAI/Orion-14B-Base", - "downloads": 1717, - "description": "Orion-14B🌐English | 🇨🇳中文 | 🇯🇵日本語 |🇰🇷한국어🤗", + "url": "https://huggingface.co/kotoba-tech/kotoba-whisper-v1.0", + "downloads": 1586, + "description": "Kotoba-WhisperKotoba-Whisper is a collection of distilled Whisper models for Japanese ASR, developed through the collaboration bewteenAsahi Ushio and Kotoba Technologies.", "source": "Hugging Face", - "score": 0.00023861641945014273, - "project_name": "Orion-14B-Base" + "score": 0.00023730031836545864, + "project_name": "kotoba-whisper-v1.0" }, { - "url": "https://huggingface.co/QuantFactory/shisa-7b-v1-GGUF", - "downloads": 1715, - "description": "QuantFactory/shisa-7b-v1-GGUFThis is quantized version of augmxnt/shisa-base-7b-v1 created using llama.cppModel Descriptionshisa-base-7b-v1 takes Mistral 7B and adds an additional 8B tokens of primarily Japanese pre-training.", + "url": "https://huggingface.co/megagonlabs/t5-base-japanese-web", + "downloads": 1544, + "description": "t5-base-japanese-web (with Byte-fallback, 32K)Descriptionmegagonlabs/t5-base-japanese-web is a T5 (Text-to-Text Transfer Transformer) model pre-trained on Japanese web texts.", "source": "Hugging Face", - "score": 0.00023833847370820894, - "project_name": "shisa-7b-v1-GGUF" + "score": 0.00023101619896359908, + "project_name": "t5-base-japanese-web" }, { - "url": "https://huggingface.co/line-corporation/japanese-large-lm-3.6b", - "downloads": 1696, - "description": "japanese-large-lm-3.6bThis repository provides a 3.6B parameters Japanese language model, trained by LINE Corporation.", + "url": "https://huggingface.co/NikolayKozloff/h2o-Llama-3-8B-Japanese-Instruct-Q8_0-GGUF", + "downloads": 1535, + "description": "NikolayKozloff/h2o-Llama-3-8B-Japanese-Instruct-Q8_0-GGUFThis model was converted to GGUF format from haqishen/h2o-Llama-3-8B-Japanese-Instruct using llama.cpp via the ggml.ai's GGUF-my-repo space.", "source": "Hugging Face", - "score": 0.00023569798915983812, - "project_name": "japanese-large-lm-3.6b" + "score": 0.0002296696019489149, + "project_name": "h2o-Llama-3-8B-Japanese-Instruct-Q8_0-GGUF" }, { "url": "https://huggingface.co/sonoisa/t5-base-japanese-question-generation", - "downloads": 1685, + "downloads": 1502, "description": "回答と回答が出てくるパラグラフを与えると質問文を生成するモデルSEE: https://github.com/sonoisa/deep-question-generation本モデルの作成ステップ概要SQuAD 1.1を日本語に機械翻訳し、不正なデータをクレンジング（有効なデータは約半分）。", "source": "Hugging Face", - "score": 0.00023416928757920237, + "score": 0.00022473207956173952, "project_name": "t5-base-japanese-question-generation" }, { - "url": "https://huggingface.co/line-corporation/japanese-large-lm-1.7b", - "downloads": 1675, - "description": "japanese-large-lm-1.7bThis repository provides a 1.7B parameters Japanese language model, trained by LINE Corporation.", + "url": "https://huggingface.co/Aratako/Ninja-v1-RP-expressive-breadcrumbs-GGUF", + "downloads": 1483, + "description": "Ninja-v1-RP-expressive-GGUF概要Aratako/Ninja-v1-RP-expressive-breadcrumbsの量子化済みGGUF版です。", "source": "Hugging Face", - "score": 0.0002327795588695335, - "project_name": "japanese-large-lm-1.7b" + "score": 0.0002218892636418507, + "project_name": "Ninja-v1-RP-expressive-breadcrumbs-GGUF" }, { - "url": "https://huggingface.co/nlp-waseda/roberta-base-japanese", - "downloads": 1671, - "description": "nlp-waseda/roberta-base-japaneseModel descriptionThis is a Japanese RoBERTa base model pretrained on Japanese Wikipedia and the Japanese portion of CC-100.How to useYou can use this model for masked language modeling as follows:from transformers import AutoTokenizer, AutoModelForMaskedLMtokenizer = AutoTokenizer.from_pretrained(\"nlp-waseda/roberta-base-japanese\")", + "url": "https://huggingface.co/lmg-anon/vntl-llama3-8b-gguf", + "downloads": 1481, + "description": "This repository contains some GGUF quantizations of the merge of the VNTL LLaMA 3 8B qlora.", "source": "Hugging Face", - "score": 0.00023222366738566598, - "project_name": "roberta-base-japanese" + "score": 0.00022159001986080974, + "project_name": "vntl-llama3-8b-gguf" }, { - "url": "https://huggingface.co/Aratako/c4ai-command-r-v01-japanese-instruct", - "downloads": 1634, - "description": "c4ai-command-r-v01-japanese-instructGGUF版はこちら/Click here for the GGUF version概要CohereForAI/c4ai-command-r-v01を、ichikara-instructionを使って追加で日本語インストラクションチューニングを施したモデルです。", + "url": "https://huggingface.co/TKU410410103/hubert-base-japanese-asr", + "downloads": 1449, + "description": "hubert-base-asrThis model is a fine-tuned version of rinna/japanese-hubert-base on the common_voice_11_0 dataset for ASR tasks.", "source": "Hugging Face", - "score": 0.00022708167115989122, - "project_name": "c4ai-command-r-v01-japanese-instruct" + "score": 0.00021680211936415483, + "project_name": "hubert-base-japanese-asr" }, { - "url": "https://huggingface.co/TKU410410103/wav2vec2-base-japanese-asr", - "downloads": 1630, - "description": "wav2vec2-base-asrThis model is a fine-tuned version of rinna/japanese-wav2vec2-base on the common_voice_11_0 dataset for ASR tasks.", + "url": "https://huggingface.co/cheonboy/sentence_embedding_japanese", + "downloads": 1426, + "description": "This is a Japanese sentence-LUKE model.", "source": "Hugging Face", - "score": 0.00022652577967602366, - "project_name": "wav2vec2-base-japanese-asr" + "score": 0.00021336081588218414, + "project_name": "sentence_embedding_japanese" }, { - "url": "https://huggingface.co/mmnga/Fugaku-LLM-13B-instruct-gguf", - "downloads": 1562, - "description": "Fugaku-LLM-13B-instruct-ggufFugaku-LLMさんが公開しているFugaku-LLM-13B-instructのggufフォーマット変換版です。", + "url": "https://huggingface.co/OrionStarAI/Orion-14B-Base", + "downloads": 1416, + "description": "Orion-14B🌐English | 🇨🇳中文 | 🇯🇵日本語 |🇰🇷한국어🤗", "source": "Hugging Face", - "score": 0.00021707562445027543, - "project_name": "Fugaku-LLM-13B-instruct-gguf" + "score": 0.00021186459697697946, + "project_name": "Orion-14B-Base" }, { - "url": "https://huggingface.co/kotoba-tech/kotoba-whisper-v1.0", - "downloads": 1546, - "description": "Kotoba-WhisperKotoba-Whisper is a collection of distilled Whisper models for Japanese ASR, developed through the collaboration bewteenAsahi Ushio and Kotoba Technologies.", + "url": "https://huggingface.co/Fugaku-LLM/Fugaku-LLM-13B-instruct", + "downloads": 1403, + "description": "Fugaku-LLM利用規約この利用規約（以下「本規約」といいます）は、富士通株式会社、国立研究開発法人理化学研究所、国立大学法人東京工業大学、国立大学法人東北大学、株式会社サイバーエージェント、国立大学法人東海国立大学機構、及び株式会社Kotoba Technologies Japan (以下「開発者」といいます)による、スーパーコンピュータ「富岳」政策対応枠における大規模言語モデル分散並列学習手法の開発の成果物として公開する大規模言語モデル（以下「Fugaku-LLM」といいます）の利用に関する条件を定めるものです。", "source": "Hugging Face", - "score": 0.00021485205851480527, - "project_name": "kotoba-whisper-v1.0" + "score": 0.0002099195124002134, + "project_name": "Fugaku-LLM-13B-instruct" }, { "url": "https://huggingface.co/tokyotech-llm/Swallow-MX-8x7b-NVE-v0.1", - "downloads": 1532, + "downloads": 1394, "description": "Swallow-MX-8x7b-NVE-v0.1Our Swallow-MX-8x7b-NVE-v0.1 model has undergone continuous pre-training from the Mixtral-8x7B-Instruct-v0.1, primarily with the addition of Japanese language data.", "source": "Hugging Face", - "score": 0.00021290643832126886, + "score": 0.00020857291538552923, "project_name": "Swallow-MX-8x7b-NVE-v0.1" }, { - "url": "https://huggingface.co/megagonlabs/t5-base-japanese-web", - "downloads": 1468, - "description": "t5-base-japanese-web (with Byte-fallback, 32K)Descriptionmegagonlabs/t5-base-japanese-web is a T5 (Text-to-Text Transfer Transformer) model pre-trained on Japanese web texts.", - "source": "Hugging Face", - "score": 0.0002040121745793882, - "project_name": "t5-base-japanese-web" - }, - { - "url": "https://huggingface.co/ku-nlp/deberta-v2-tiny-japanese-char-wwm", - "downloads": 1445, - "description": "Model Card for Japanese character-level DeBERTa V2 tinyModel descriptionThis is a Japanese DeBERTa V2 tiny model pre-trained on Japanese Wikipedia, the Japanese portion of CC-100, and the Japanese portion of OSCAR.This model is trained with character-level tokenization and whole word masking.", - "source": "Hugging Face", - "score": 0.0002008157985471498, - "project_name": "deberta-v2-tiny-japanese-char-wwm" - }, - { - "url": "https://huggingface.co/Lasorco/lametta_old", - "downloads": 1406, - "description": "old？", + "url": "https://huggingface.co/mmnga/Fugaku-LLM-13B-instruct-gguf", + "downloads": 1384, + "description": "Fugaku-LLM-13B-instruct-ggufFugaku-LLMさんが公開しているFugaku-LLM-13B-instructのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 0.00019539585657944128, - "project_name": "lametta_old" + "score": 0.00020707669648032458, + "project_name": "Fugaku-LLM-13B-instruct-gguf" }, { - "url": "https://huggingface.co/dahara1/weblab-10b-instruction-sft-GPTQ", - "downloads": 1397, - "description": "weblab-10b-instruction-sft-GPTQOriginal model weblab-10b-instruction-sft which is a Japanese-centric multilingual GPT-NeoX model of 10 billion parameters created by matsuo-labTakeshi Kojima.", + "url": "https://huggingface.co/elyza/ELYZA-japanese-CodeLlama-7b-instruct", + "downloads": 1323, + "description": "ELYZA-japanese-CodeLlama-7bModel DescriptionELYZA-japanese-CodeLlama-7b は、 Code Llamaをベースとして日本語能力を拡張するために追加事前学習を行ったモデルです。", "source": "Hugging Face", - "score": 0.0001941451007407393, - "project_name": "weblab-10b-instruction-sft-GPTQ" + "score": 0.00019794976115857616, + "project_name": "ELYZA-japanese-CodeLlama-7b-instruct" }, { - "url": "https://huggingface.co/TKU410410103/hubert-base-japanese-asr", - "downloads": 1371, - "description": "hubert-base-asrThis model is a fine-tuned version of rinna/japanese-hubert-base on the common_voice_11_0 dataset for ASR tasks.", + "url": "https://huggingface.co/ku-nlp/deberta-v2-tiny-japanese", + "downloads": 1321, + "description": "Model Card for Japanese DeBERTa V2 tinyModel descriptionThis is a Japanese DeBERTa V2 tiny model pre-trained on Japanese Wikipedia, the Japanese portion of CC-100, and the Japanese portion of OSCAR.How to useYou can use this model for masked language modeling as follows:from transformers import AutoTokenizer, AutoModelForMaskedLMtokenizer = AutoTokenizer.from_pretrained('ku-nlp/deberta-v2-tiny-japanese')", "source": "Hugging Face", - "score": 0.00019053180609560026, - "project_name": "hubert-base-japanese-asr" + "score": 0.00019765051737753524, + "project_name": "deberta-v2-tiny-japanese" }, { - "url": "https://huggingface.co/elyza/ELYZA-japanese-CodeLlama-7b-instruct", - "downloads": 1364, - "description": "ELYZA-japanese-CodeLlama-7bModel DescriptionELYZA-japanese-CodeLlama-7b は、 Code Llamaをベースとして日本語能力を拡張するために追加事前学習を行ったモデルです。", + "url": "https://huggingface.co/izumi-lab/electra-base-japanese-generator", + "downloads": 1316, + "description": "ELECTRA base Japanese generatorThis is a ELECTRA model pretrained on texts in the Japanese language.", "source": "Hugging Face", - "score": 0.00018955899599883206, - "project_name": "ELYZA-japanese-CodeLlama-7b-instruct" + "score": 0.0001969024079249329, + "project_name": "electra-base-japanese-generator" }, { - "url": "https://huggingface.co/MCZK/ArrowPro-7B-KUJIRA-GGUF", - "downloads": 1357, - "description": "DataPilot様の ArrowPro-7B-KUJIRA をGGUF形式に変換したものです。", + "url": "https://huggingface.co/dahara1/weblab-10b-instruction-sft-GPTQ", + "downloads": 1266, + "description": "weblab-10b-instruction-sft-GPTQOriginal model weblab-10b-instruction-sft which is a Japanese-centric multilingual GPT-NeoX model of 10 billion parameters created by matsuo-labTakeshi Kojima.", "source": "Hugging Face", - "score": 0.00018858618590206387, - "project_name": "ArrowPro-7B-KUJIRA-GGUF" + "score": 0.0001894213133989096, + "project_name": "weblab-10b-instruction-sft-GPTQ" }, { "url": "https://huggingface.co/stabilityai/japanese-stablelm-base-beta-70b", - "downloads": 1335, + "downloads": 1246, "description": "Japanese-StableLM-Base-Beta-70BA cute robot wearing a kimono writes calligraphy with one single brush — Stable Diffusion XLModel Descriptionjapanese-stablelm-base-beta-70b is a 70B-parameter decoder-only language model based on Llama-2-70b that has been fine-tuned on a diverse collection of Japanese data, with the intent of maximizing downstream performance on Japanese language tasks.", "source": "Hugging Face", - "score": 0.0001855287827407924, + "score": 0.0001864288755885003, "project_name": "japanese-stablelm-base-beta-70b" }, { - "url": "https://huggingface.co/mmnga/Ninja-v1-gguf", - "downloads": 1318, - "description": "Ninja-v1-ggufLocal-Novel-LLM-projectさんが公開しているNinja-v1のggufフォーマット変換版です。", + "url": "https://huggingface.co/TKU410410103/uniTKU-hubert-japanese-asr", + "downloads": 1207, + "description": "uniTKU-hubert-japanese-asrThis model was fine-tuned on a dataset provided by uniTKU, and it has maintained the original performance metrics on the common_voice_11_0 dataset.", "source": "Hugging Face", - "score": 0.00018316624393435534, - "project_name": "Ninja-v1-gguf" + "score": 0.00018059362185820214, + "project_name": "uniTKU-hubert-japanese-asr" }, { - "url": "https://huggingface.co/mmnga/DataPilot-ArrowPro-7B-KUJIRA-gguf", - "downloads": 1318, - "description": "DataPilot-ArrowPro-7B-KUJIRA-ggufDataPilotさんが公開しているArrowPro-7B-KUJIRAのggufフォーマット変換版です。", + "url": "https://huggingface.co/tokyotech-llm/Swallow-13b-instruct-v0.1", + "downloads": 1206, + "description": "SwallowOur Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data.", "source": "Hugging Face", - "score": 0.00018316624393435534, - "project_name": "DataPilot-ArrowPro-7B-KUJIRA-gguf" + "score": 0.00018044399996768168, + "project_name": "Swallow-13b-instruct-v0.1" }, { "url": "https://huggingface.co/KBlueLeaf/guanaco-7b-leh-v2", - "downloads": 1299, + "downloads": 1202, "description": "Guanaco-leh-V2: A Multilingual Instruction-Following Language Model Based on LLaMA", "source": "Hugging Face", - "score": 0.0001805257593859845, + "score": 0.0001798455124055998, "project_name": "guanaco-7b-leh-v2" }, + { + "url": "https://huggingface.co/Aratako/c4ai-command-r-v01-japanese-instruct", + "downloads": 1195, + "description": "c4ai-command-r-v01-japanese-instructGGUF版はこちら/Click here for the GGUF version概要CohereForAI/c4ai-command-r-v01を、ichikara-instructionを使って追加で日本語インストラクションチューニングを施したモデルです。", + "source": "Hugging Face", + "score": 0.00017879815917195654, + "project_name": "c4ai-command-r-v01-japanese-instruct" + }, { "url": "https://huggingface.co/dahara1/ELYZA-japanese-Llama-2-7b-fast-instruct-GPTQ", - "downloads": 1292, + "downloads": 1195, "description": "Model Card for Model IDOriginal model elyza/ELYZA-japanese-Llama-2-7b-fast-instruct which is based on Meta's \"Llama 2\" and has undergone additional pre-training in Japanese, and thier original post-training and speed up tuning.", "source": "Hugging Face", - "score": 0.0001795529492892163, + "score": 0.00017879815917195654, "project_name": "ELYZA-japanese-Llama-2-7b-fast-instruct-GPTQ" }, { - "url": "https://huggingface.co/tokyotech-llm/Swallow-13b-instruct-v0.1", - "downloads": 1286, + "url": "https://huggingface.co/TKU410410103/hubert-large-japanese-asr", + "downloads": 1182, + "description": "hubert-large-asrThis model is a fine-tuned version of rinna/japanese-hubert-large ASR.", + "source": "Hugging Face", + "score": 0.0001768530745951905, + "project_name": "hubert-large-japanese-asr" + }, + { + "url": "https://huggingface.co/tokyotech-llm/Swallow-13b-hf", + "downloads": 1117, "description": "SwallowOur Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data.", "source": "Hugging Face", - "score": 0.00017871911206341498, - "project_name": "Swallow-13b-instruct-v0.1" + "score": 0.0001671276517113602, + "project_name": "Swallow-13b-hf" }, { - "url": "https://huggingface.co/cheonboy/sentence_embedding_japanese", - "downloads": 1269, - "description": "This is a Japanese sentence-LUKE model.", + "url": "https://huggingface.co/hotchpotch/japanese-bge-reranker-v2-m3-v1", + "downloads": 1110, + "description": "hotchpotch/japanese-bge-reranker-v2-m3-v1日本語で学習させた Reranker (CrossEncoder) シリーズです。", "source": "Hugging Face", - "score": 0.00017635657325697793, - "project_name": "sentence_embedding_japanese" + "score": 0.00016608029847771697, + "project_name": "japanese-bge-reranker-v2-m3-v1" }, { - "url": "https://huggingface.co/mmnga/ELYZA-japanese-Llama-2-7b-fast-gguf", - "downloads": 1241, - "description": "ELYZA-japanese-Llama-2-7b-fast-ggufELYZAさんが公開しているELYZA-japanese-Llama-2-7b-fastのggufフォーマット変換版です。", + "url": "https://huggingface.co/Lasorco/lametta_old", + "downloads": 1088, + "description": "old？", "source": "Hugging Face", - "score": 0.00017246533286990512, - "project_name": "ELYZA-japanese-Llama-2-7b-fast-gguf" + "score": 0.0001627886168862667, + "project_name": "lametta_old" }, { - "url": "https://huggingface.co/mmnga/rinna-llama-3-youko-8b-gguf", - "downloads": 1177, - "description": "rinna-llama-3-youko-8b-ggufrinnaさんが公開しているllama-3-youko-8bのggufフォーマット変換版です。", + "url": "https://huggingface.co/umiyuki/Umievo-itr012-Gleipnir-7B", + "downloads": 1081, + "description": "Umievo-itr012-Gleipnir-7Bこのモデルは強力な４つの日本語モデルを進化的アルゴリズムで進化的マージしたものです。", "source": "Hugging Face", - "score": 0.00016357106912802445, - "project_name": "rinna-llama-3-youko-8b-gguf" + "score": 0.00016174126365262346, + "project_name": "Umievo-itr012-Gleipnir-7B" }, { - "url": "https://huggingface.co/TKU410410103/uniTKU-hubert-japanese-asr", - "downloads": 1153, - "description": "uniTKU-hubert-japanese-asrThis model was fine-tuned on a dataset provided by uniTKU, and it has maintained the original performance metrics on the common_voice_11_0 dataset.", + "url": "https://huggingface.co/mmnga/ELYZA-japanese-Llama-2-7b-fast-gguf", + "downloads": 1055, + "description": "ELYZA-japanese-Llama-2-7b-fast-ggufELYZAさんが公開しているELYZA-japanese-Llama-2-7b-fastのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 0.00016023572022481918, - "project_name": "uniTKU-hubert-japanese-asr" + "score": 0.00015785109449909134, + "project_name": "ELYZA-japanese-Llama-2-7b-fast-gguf" }, { - "url": "https://huggingface.co/mmnga/c4ai-command-r-plus-gguf", - "downloads": 1118, - "description": "c4ai-command-r-plus-ggufCohereForAIさんが公開しているc4ai-command-r-plusのggufフォーマット変換版です。", + "url": "https://huggingface.co/nitky/Oumuamua-7b-base", + "downloads": 1031, + "description": "Oumuamua-7b-baseThis is a merge of pre-trained language models created using mergekit.", "source": "Hugging Face", - "score": 0.00015537166974097819, - "project_name": "c4ai-command-r-plus-gguf" + "score": 0.00015426016912660018, + "project_name": "Oumuamua-7b-base" }, { - "url": "https://huggingface.co/rinna/japanese-hubert-large", - "downloads": 1116, - "description": "rinna/japanese-hubert-largeOverviewThis is a Japanese HuBERT Large model trained by rinna Co.", + "url": "https://huggingface.co/skytnt/gpt2-japanese-lyric-small", + "downloads": 1026, + "description": "Japanese GPT2 Lyric ModelModel descriptionThe model is used to generate Japanese lyrics.", "source": "Hugging Face", - "score": 0.00015509372399904442, - "project_name": "japanese-hubert-large" + "score": 0.00015351205967399783, + "project_name": "gpt2-japanese-lyric-small" }, { - "url": "https://huggingface.co/TKU410410103/hubert-large-japanese-asr", - "downloads": 1097, - "description": "hubert-large-asrThis model is a fine-tuned version of rinna/japanese-hubert-large ASR.", + "url": "https://huggingface.co/esnya/japanese_speecht5_tts", + "downloads": 998, + "description": "SpeechT5 (TTS task) for JapaneseSpeechT5 model fine-tuned for Japanese speech synthesis (text-to-speech)", "source": "Hugging Face", - "score": 0.0001524532394506736, - "project_name": "hubert-large-japanese-asr" + "score": 0.0001493226467394248, + "project_name": "japanese_speecht5_tts" }, { "url": "https://huggingface.co/OrionStarAI/Orion-14B-Chat-RAG", - "downloads": 1094, + "downloads": 997, "description": "Orion-14B🌐English | 🇨🇳中文 | 🇯🇵日本語 | 🇰🇷한국어🤗", "source": "Hugging Face", - "score": 0.00015203632083777294, + "score": 0.00014917302484890432, "project_name": "Orion-14B-Chat-RAG" }, { - "url": "https://huggingface.co/mmnga/ELYZA-japanese-Llama-2-7b-gguf", - "downloads": 1071, - "description": "ELYZA-japanese-Llama-2-7b-ggufELYZAさんが公開しているELYZA-japanese-Llama-2-7bのggufフォーマット変換版です。", + "url": "https://huggingface.co/stabilityai/japanese-stablelm-2-instruct-1_6b", + "downloads": 981, + "description": "By clicking \"Agree\", you agree to the License Agreement and acknowledge Stability AI's Privacy Policy.", "source": "Hugging Face", - "score": 0.00014883994480553457, - "project_name": "ELYZA-japanese-Llama-2-7b-gguf" + "score": 0.0001467790746005769, + "project_name": "japanese-stablelm-2-instruct-1_6b" }, { - "url": "https://huggingface.co/mmnga/haqishen-Llama-3-8B-Japanese-Instruct-gguf", - "downloads": 1060, - "description": "haqishen-Llama-3-8B-Japanese-Instruct-ggufhaqishenさんが公開しているLlama-3-8B-Japanese-Instructのggufフォーマット変換版です。", + "url": "https://huggingface.co/mmnga/rinna-llama-3-youko-8b-gguf", + "downloads": 976, + "description": "rinna-llama-3-youko-8b-ggufrinnaさんが公開しているllama-3-youko-8bのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 0.0001473112432248988, - "project_name": "haqishen-Llama-3-8B-Japanese-Instruct-gguf" + "score": 0.00014603096514797454, + "project_name": "rinna-llama-3-youko-8b-gguf" }, { - "url": "https://huggingface.co/hotchpotch/japanese-bge-reranker-v2-m3-v1", - "downloads": 1046, - "description": "hotchpotch/japanese-bge-reranker-v2-m3-v1日本語で学習させた Reranker (CrossEncoder) シリーズです。", + "url": "https://huggingface.co/mmnga/Ninja-v1-gguf", + "downloads": 935, + "description": "Ninja-v1-ggufLocal-Novel-LLM-projectさんが公開しているNinja-v1のggufフォーマット変換版です。", "source": "Hugging Face", - "score": 0.00014536562303136242, - "project_name": "japanese-bge-reranker-v2-m3-v1" + "score": 0.00013989646763663545, + "project_name": "Ninja-v1-gguf" }, { - "url": "https://huggingface.co/tokyotech-llm/Swallow-13b-hf", - "downloads": 1042, - "description": "SwallowOur Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data.", - "source": "Hugging Face", - "score": 0.0001448097315474949, - "project_name": "Swallow-13b-hf" + "url": "https://huggingface.co/stabilityai/japanese-stablelm-2-base-1_6b", + "downloads": 927, + "description": "By clicking \"Agree\", you agree to the License Agreement and acknowledge Stability AI's Privacy Policy.", + "source": "Hugging Face", + "score": 0.00013869949251247175, + "project_name": "japanese-stablelm-2-base-1_6b" }, { - "url": "https://huggingface.co/skytnt/gpt2-japanese-lyric-small", - "downloads": 1041, - "description": "Japanese GPT2 Lyric ModelModel descriptionThe model is used to generate Japanese lyrics.", + "url": "https://huggingface.co/karakuri-ai/karakuri-lm-70b-v0.1", + "downloads": 927, + "description": "KARAKURI LMKARAKURI LM is a pretrained language model that builds upon Llama 2.Our model enhances Llama 2's capabilities by incorporating additional Japanese vocabulary and further pretraining on a mixture of Japanese and multilingual corpora.", "source": "Hugging Face", - "score": 0.000144670758676528, - "project_name": "gpt2-japanese-lyric-small" + "score": 0.00013869949251247175, + "project_name": "karakuri-lm-70b-v0.1" }, { - "url": "https://huggingface.co/nitky/Oumuamua-7b-base", - "downloads": 1031, - "description": "Oumuamua-7b-baseThis is a merge of pre-trained language models created using mergekit.", + "url": "https://huggingface.co/OrionStarAI/Orion-14B-Chat", + "downloads": 925, + "description": "Orion-14B🌐English | 🇨🇳中文 | 🇯🇵日本語 | 🇰🇷한국어🤗", "source": "Hugging Face", - "score": 0.00014328102996685914, - "project_name": "Oumuamua-7b-base" + "score": 0.0001384002487314308, + "project_name": "Orion-14B-Chat" }, { - "url": "https://huggingface.co/stabilityai/japanese-stablelm-2-base-1_6b", - "downloads": 1030, - "description": "By clicking \"Agree\", you agree to the License Agreement and acknowledge Stability AI's Privacy Policy.", + "url": "https://huggingface.co/Fugaku-LLM/Fugaku-LLM-13B", + "downloads": 911, + "description": "Fugaku-LLM利用規約この利用規約（以下「本規約」といいます）は、富士通株式会社、国立研究開発法人理化学研究所、国立大学法人東京工業大学、国立大学法人東北大学、株式会社サイバーエージェント、国立大学法人東海国立大学機構、及び株式会社Kotoba Technologies Japan (以下「開発者」といいます)による、スーパーコンピュータ「富岳」政策対応枠における大規模言語モデル分散並列学習手法の開発の成果物として公開する大規模言語モデル（以下「Fugaku-LLM」といいます）の利用に関する条件を定めるものです。", "source": "Hugging Face", - "score": 0.00014314205709589224, - "project_name": "japanese-stablelm-2-base-1_6b" + "score": 0.0001363055422641443, + "project_name": "Fugaku-LLM-13B" }, { - "url": "https://huggingface.co/mmnga/DataPilot-ArrowPro-7B-RobinHood-gguf", - "downloads": 1026, - "description": "DataPilot-ArrowPro-7B-RobinHood-ggufDataPilotさんが公開しているArrowPro-7B-RobinHoodのggufフォーマット変換版です。", + "url": "https://huggingface.co/mmnga/ELYZA-japanese-Llama-2-7b-gguf", + "downloads": 896, + "description": "ELYZA-japanese-Llama-2-7b-ggufELYZAさんが公開しているELYZA-japanese-Llama-2-7bのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 0.0001425861656120247, - "project_name": "DataPilot-ArrowPro-7B-RobinHood-gguf" + "score": 0.0001340612139063373, + "project_name": "ELYZA-japanese-Llama-2-7b-gguf" }, { - "url": "https://huggingface.co/umiyuki/Umievo-itr012-Gleipnir-7B", - "downloads": 1025, - "description": "Umievo-itr012-Gleipnir-7Bこのモデルは強力な４つの日本語モデルを進化的アルゴリズムで進化的マージしたものです。", + "url": "https://huggingface.co/line-corporation/japanese-large-lm-3.6b-instruction-sft", + "downloads": 893, + "description": "japanese-large-lm-3.6b-instruction-sftThis repository provides a 3.6B parameters Japanese language model, fine-tuned and trained by LINE Corporation.", "source": "Hugging Face", - "score": 0.00014244719274105781, - "project_name": "Umievo-itr012-Gleipnir-7B" + "score": 0.0001336123482347759, + "project_name": "japanese-large-lm-3.6b-instruction-sft" }, { - "url": "https://huggingface.co/mmnga/lightblue-suzume-llama-3-8B-multilingual-gguf", - "downloads": 998, - "description": "lightblue-suzume-llama-3-8B-multilingual-gguflightblueさんが公開しているsuzume-llama-3-8B-multilingualのggufフォーマット変換版です。", + "url": "https://huggingface.co/mmnga/aya-23-35B-gguf", + "downloads": 874, + "description": "aya-23-35B-ggufCohereForAIさんが公開しているaya-23-35Bのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 0.0001386949252249519, - "project_name": "lightblue-suzume-llama-3-8B-multilingual-gguf" + "score": 0.00013076953231488706, + "project_name": "aya-23-35B-gguf" }, { - "url": "https://huggingface.co/ku-nlp/deberta-v3-base-japanese", - "downloads": 995, - "description": "Model Card for Japanese DeBERTa V3 baseModel", + "url": "https://huggingface.co/stabilityai/japanese-stablelm-base-alpha-7b", + "downloads": 867, + "description": "Japanese-StableLM-Base-Alpha-7B\"A parrot able to speak Japanese, ukiyoe, edo period\" — Stable Diffusion XLModel Descriptionjapanese-stablelm-base-alpha-7b is a 7B-parameter decoder-only language model pre-trained on a diverse collection of Japanese and English datasets which focus on maximizing Japanese language modeling performance and Japanese downstream task performance.", "source": "Hugging Face", - "score": 0.00013827800661205125, - "project_name": "deberta-v3-base-japanese" + "score": 0.0001297221790812438, + "project_name": "japanese-stablelm-base-alpha-7b" }, { "url": "https://huggingface.co/studio-ousia/luke-japanese-large", - "downloads": 986, + "downloads": 862, "description": "luke-japanese-largeluke-japanese is the Japanese version of LUKE (LanguageUnderstanding with Knowledge-based Embeddings), a pre-trainedknowledge-enhanced contextualized representation of words and entities.", "source": "Hugging Face", - "score": 0.00013702725077334929, + "score": 0.00012897406962864146, "project_name": "luke-japanese-large" }, { - "url": "https://huggingface.co/stabilityai/japanese-stablelm-2-instruct-1_6b", - "downloads": 983, - "description": "By clicking \"Agree\", you agree to the License Agreement and acknowledge Stability AI's Privacy Policy.", - "source": "Hugging Face", - "score": 0.00013661033216044862, - "project_name": "japanese-stablelm-2-instruct-1_6b" - }, - { - "url": "https://huggingface.co/Fugaku-LLM/Fugaku-LLM-13B", - "downloads": 980, - "description": "Fugaku-LLM利用規約この利用規約（以下「本規約」といいます）は、富士通株式会社、国立研究開発法人理化学研究所、国立大学法人東京工業大学、国立大学法人東北大学、株式会社サイバーエージェント、国立大学法人東海国立大学機構、及び株式会社Kotoba Technologies Japan (以下「開発者」といいます)による、スーパーコンピュータ「富岳」政策対応枠における大規模言語��デル分散並列学習手法の開発の成果物として公開する大規模言語モデル（以下「Fugaku-LLM」といいます）の利用に関する条件を定めるものです。", - "source": "Hugging Face", - "score": 0.00013619341354754796, - "project_name": "Fugaku-LLM-13B" - }, - { - "url": "https://huggingface.co/mmnga/umiyuki-Japanese-Chat-Umievo-itr001-7b-gguf", - "downloads": 971, - "description": "umiyuki-Japanese-Chat-Umievo-itr001-7b-ggufumiyukiさんが公開しているJapanese-Chat-Umievo-itr001-7bのggufフォーマット変換版です。", + "url": "https://huggingface.co/jurabi/bert-ner-japanese", + "downloads": 857, + "description": "BERTによる日本語固有表現抽出のモデルBertForTokenClassificationを用いて、日本語の文から固有表現を抽出します。", "source": "Hugging Face", - "score": 0.000134942657708846, - "project_name": "umiyuki-Japanese-Chat-Umievo-itr001-7b-gguf" + "score": 0.00012822596017603914, + "project_name": "bert-ner-japanese" }, { - "url": "https://huggingface.co/ku-nlp/deberta-v2-tiny-japanese", - "downloads": 963, - "description": "Model Card for Japanese DeBERTa V2 tinyModel descriptionThis is a Japanese DeBERTa V2 tiny model pre-trained on Japanese Wikipedia, the Japanese portion of CC-100, and the Japanese portion of OSCAR.How to useYou can use this model for masked language modeling as follows:from transformers import AutoTokenizer, AutoModelForMaskedLMtokenizer = AutoTokenizer.from_pretrained('ku-nlp/deberta-v2-tiny-japanese')", + "url": "https://huggingface.co/mmnga/DataPilot-ArrowPro-7B-KUJIRA-gguf", + "downloads": 852, + "description": "DataPilot-ArrowPro-7B-KUJIRA-ggufDataPilotさんが公開しているArrowPro-7B-KUJIRAのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 0.0001338308747411109, - "project_name": "deberta-v2-tiny-japanese" + "score": 0.0001274778507234368, + "project_name": "DataPilot-ArrowPro-7B-KUJIRA-gguf" }, { - "url": "https://huggingface.co/line-corporation/japanese-large-lm-3.6b-instruction-sft", - "downloads": 958, - "description": "japanese-large-lm-3.6b-instruction-sftThis repository provides a 3.6B parameters Japanese language model, fine-tuned and trained by LINE Corporation.", + "url": "https://huggingface.co/ku-nlp/deberta-v3-base-japanese", + "downloads": 831, + "description": "Model Card for Japanese DeBERTa V3 baseModel", "source": "Hugging Face", - "score": 0.00013313601038627648, - "project_name": "japanese-large-lm-3.6b-instruction-sft" + "score": 0.000124335791022507, + "project_name": "deberta-v3-base-japanese" }, { - "url": "https://huggingface.co/karakuri-ai/karakuri-lm-70b-chat-v0.1", - "downloads": 957, - "description": "KARAKURI LMKARAKURI LM is a pretrained language model that builds upon Llama 2.Our model enhances Llama 2's capabilities by incorporating additional Japanese vocabulary and further pretraining on a mixture of Japanese and multilingual corpora.", + "url": "https://huggingface.co/Mizuiro-sakura/luke-japanese-base-finetuned-ner", + "downloads": 827, + "description": "このモデルはluke-japanese-baseをファインチューニングして、固有表現抽出（NER）に用いれるようにしたものです。", "source": "Hugging Face", - "score": 0.0001329970375153096, - "project_name": "karakuri-lm-70b-chat-v0.1" + "score": 0.00012373730346042515, + "project_name": "luke-japanese-base-finetuned-ner" }, { - "url": "https://huggingface.co/mmnga/ryota39-Phi-3-mini-4k-instruct-dpo-gguf", - "downloads": 955, - "description": "ryota39-Phi-3-mini-4k-instruct-dpo-ggufryota39さんが公開しているPhi-3-mini-4k-instruct-dpoのggufフォーマット変換版です。", + "url": "https://huggingface.co/stabilityai/japanese-stablelm-instruct-alpha-7b-v2", + "downloads": 808, + "description": "Japanese-StableLM-Instruct-Alpha-7B-v2\"A parrot able to speak Japanese, ukiyoe, edo period\" — Stable Diffusion XLModel Descriptionjapanese-stablelm-instruct-alpha-7b-v2 is a 7B parameter decoder-only language models pre-trained built on top of the Japanese-StableLM-Base-Alpha-7B model and further fine-tuned on various instruction-following datasets.", "source": "Hugging Face", - "score": 0.00013271909177337582, - "project_name": "ryota39-Phi-3-mini-4k-instruct-dpo-gguf" + "score": 0.00012089448754053632, + "project_name": "japanese-stablelm-instruct-alpha-7b-v2" }, { - "url": "https://huggingface.co/ken11/albert-base-japanese-v1", - "downloads": 945, - "description": "albert-base-japanese-v1日本語事前学習済みALBERTモデルですHow to useファインチュー��ングこのモデルはPreTrainedモデルです基本的には各種タスク用にファインチューニングして使用されることを想定していますFill-MaskこのモデルではTokenizerにSentencepieceを利用していますそのままでは[MASK]トークンのあとに余計なトークンが混入する問題があるので、利用する際には以下のようにする必要がありますfor PyTorchfrom transformers import (AlbertForMaskedLM, AlbertTokenizerFast)import torchtokenizer = AlbertTokenizerFast.from_pretrained(\"ken11/albert-base-japanese-v1\")", + "url": "https://huggingface.co/llm-book/bert-base-japanese-v3-marc_ja", + "downloads": 794, + "description": "bert-base-japanese-v3-marc_ja「大規模言語モデル入門」の第5章で紹介している(感情分析)のモデルです。", "source": "Hugging Face", - "score": 0.00013132936306370696, - "project_name": "albert-base-japanese-v1" + "score": 0.00011879978107324979, + "project_name": "bert-base-japanese-v3-marc_ja" }, { - "url": "https://huggingface.co/karakuri-ai/karakuri-lm-70b-v0.1", - "downloads": 937, - "description": "KARAKURI LMKARAKURI LM is a pretrained language model that builds upon Llama 2.Our model enhances Llama 2's capabilities by incorporating additional Japanese vocabulary and further pretraining on a mixture of Japanese and multilingual corpora.", + "url": "https://huggingface.co/TFMC/Japanese-Starling-ChatV-7B-GGUF", + "downloads": 773, + "description": "Japanese-Starling-ChatV-7B-GGUFGGUF conversion of \"Japanese-Starling-ChatV-7B\"\"Japanese-Starling-ChatV-7B\" is a Japanese chat model built on top of \"chatntq-ja-7b-v1.0\", originally based on Mistral-7B-v0.1.I applied the chat vector acquired by subtracting the weights of Mistral-7B-v0.1 from the weights of \"Starling-LM-7B-beta\" to this model.", "source": "Hugging Face", - "score": 0.00013021758009597187, - "project_name": "karakuri-lm-70b-v0.1" + "score": 0.00011565772137232001, + "project_name": "Japanese-Starling-ChatV-7B-GGUF" }, { - "url": "https://huggingface.co/mmnga/lightblue-suzume-llama-3-8B-japanese-gguf", - "downloads": 929, - "description": "lightblue-suzume-llama-3-8B-japanese-gguflightblueさんが公開しているsuzume-llama-3-8B-japaneseのggufフォーマット変換版です。", + "url": "https://huggingface.co/bclavie/fio-base-japanese-v0.1", + "downloads": 771, + "description": "fio-base-japanese-v0.1日本語版は近日公開予定です（日本語を勉強中なので、間違いはご容赦ください！", "source": "Hugging Face", - "score": 0.0001291057971282368, - "project_name": "lightblue-suzume-llama-3-8B-japanese-gguf" + "score": 0.00011535847759127908, + "project_name": "fio-base-japanese-v0.1" }, { - "url": "https://huggingface.co/TFMC/Japanese-Starling-ChatV-7B-GGUF", - "downloads": 914, - "description": "Japanese-Starling-ChatV-7B-GGUFGGUF conversion of \"Japanese-Starling-ChatV-7B\"\"Japanese-Starling-ChatV-7B\" is a Japanese chat model built on top of \"chatntq-ja-7b-v1.0\", originally based on Mistral-7B-v0.1.I applied the chat vector acquired by subtracting the weights of Mistral-7B-v0.1 from the weights of \"Starling-LM-7B-beta\" to this model.", + "url": "https://huggingface.co/nitky/Oumuamua-7b-instruct", + "downloads": 770, + "description": "Oumuamua-7b-instructThis is a merge of pre-trained language models created using mergekit.Output example[INST] <<SYS>>あなたは日本語を話す優秀なアシスタントです。", "source": "Hugging Face", - "score": 0.00012702120406373352, - "project_name": "Japanese-Starling-ChatV-7B-GGUF" + "score": 0.00011520885570075861, + "project_name": "Oumuamua-7b-instruct" }, { - "url": "https://huggingface.co/mmnga/tokyotech-llm-Swallow-MS-7b-instruct-v0.1-gguf", - "downloads": 900, - "description": "tokyotech-llm-Swallow-MS-7b-instruct-v0.1-gguftokyotech-llmさんが公開しているSwallow-MS-7b-instruct-v0.1のggufフォーマット変換版です。", + "url": "https://huggingface.co/abeja/gpt2-large-japanese", + "downloads": 769, + "description": "gpt2-large-japaneseThis repository provides a large sized Japanese GPT-2 model.", "source": "Hugging Face", - "score": 0.0001250755838701971, - "project_name": "tokyotech-llm-Swallow-MS-7b-instruct-v0.1-gguf" + "score": 0.00011505923381023815, + "project_name": "gpt2-large-japanese" }, { - "url": "https://huggingface.co/hajime9652/xlnet-japanese", - "downloads": 896, - "description": "XLNet-japaneseModel descriptionThis model require Mecab and senetencepiece with XLNetTokenizer.", + "url": "https://huggingface.co/KoichiYasuoka/roberta-small-japanese-luw-upos", + "downloads": 765, + "description": "roberta-small-japanese-luw-uposModel", "source": "Hugging Face", - "score": 0.00012451969238632958, - "project_name": "xlnet-japanese" + "score": 0.00011446074624815629, + "project_name": "roberta-small-japanese-luw-upos" }, { - "url": "https://huggingface.co/Local-Novel-LLM-project/Vecteus-v1", - "downloads": 895, - "description": "Our ModelsVecteusNinja-v1Ninja-v1-NSFWNinja-v1-128kNinja-v1-NSFW-128kModel Card for VecTeus-v1.0The Mistral-7B--based Large Language Model (LLM) is an noveldataset fine-tuned version of the Mistral-7B-v0.1VecTeus has the following changes compared to Mistral-7B-v0.1.128k context window (8k context in v0.1)Achieving both high quality Japanese and English generationCan be generated NSFWMemory ability that does not forget even after long-context generationThis model was created with the help of GPUs from the f", + "url": "https://huggingface.co/retrieva-jp/t5-small-short", + "downloads": 763, + "description": "Model card for model IDThis is a T5 v1.1 model, pre-trained on a Japanese corpus.", "source": "Hugging Face", - "score": 0.00012438071951536268, - "project_name": "Vecteus-v1" + "score": 0.00011416150246711535, + "project_name": "t5-small-short" }, { - "url": "https://huggingface.co/llm-book/bert-base-japanese-v3-crf-ner-wikipedia-dataset", - "downloads": 891, - "description": "llm-book/bert-base-japanese-v3-crf-ner-wikipedia-dataset「大規模言語モデル入門」の第6章で紹介している固有表現認識のモデルです。", + "url": "https://huggingface.co/mmnga/lightblue-suzume-llama-3-8B-multilingual-gguf", + "downloads": 757, + "description": "lightblue-suzume-llama-3-8B-multilingual-gguflightblueさんが公開しているsuzume-llama-3-8B-multilingualのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 0.00012382482803149515, - "project_name": "bert-base-japanese-v3-crf-ner-wikipedia-dataset" + "score": 0.00011326377112399256, + "project_name": "lightblue-suzume-llama-3-8B-multilingual-gguf" }, { - "url": "https://huggingface.co/stabilityai/japanese-stablelm-base-alpha-7b", - "downloads": 881, - "description": "Japanese-StableLM-Base-Alpha-7B\"A parrot able to speak Japanese, ukiyoe, edo period\" — Stable Diffusion XLModel Descriptionjapanese-stablelm-base-alpha-7b is a 7B-parameter decoder-only language model pre-trained on a diverse collection of Japanese and English datasets which focus on maximizing Japanese language modeling performance and Japanese downstream task performance.", + "url": "https://huggingface.co/Aratako/Ninja-v1-RP-expressive-v2-GGUF", + "downloads": 752, + "description": "Ninja-v1-RP-expressive-GGUF概要Aratako/Ninja-v1-RP-expressive-v2の量子化済みGGUF版です。", "source": "Hugging Face", - "score": 0.0001224350993218263, - "project_name": "japanese-stablelm-base-alpha-7b" + "score": 0.00011251566167139023, + "project_name": "Ninja-v1-RP-expressive-v2-GGUF" }, { "url": "https://huggingface.co/mmnga/stockmark-100b-gguf", - "downloads": 878, + "downloads": 748, "description": "stockmark-100b-ggufstockmarkさんが公開しているstockmark-100bのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 0.00012201818070892563, + "score": 0.00011191717410930837, "project_name": "stockmark-100b-gguf" }, { - "url": "https://huggingface.co/mmnga/Meta-Llama-3-8B-Instruct-gguf", - "downloads": 874, - "description": "Meta-Llama-3-8B-Instruct-ggufmeta-llamaさんが公開しているMeta-Llama-3-8B-Instructのggufフォーマット変換版です。", - "source": "Hugging Face", - "score": 0.00012146228922505808, - "project_name": "Meta-Llama-3-8B-Instruct-gguf" - }, - { - "url": "https://huggingface.co/llm-book/bert-base-japanese-v3-marc_ja", - "downloads": 869, - "description": "bert-base-japanese-v3-marc_ja「大規模言語モデル入門」の第5章で紹介している(感情分析)のモデルです。", + "url": "https://huggingface.co/stabilityai/japanese-stablelm-3b-4e1t-instruct", + "downloads": 735, + "description": "Japanese StableLM-3B-4E1T InstructModel DescriptionThis is a 3B-parameter decoder-only Japanese language model fine-tuned on instruction-following datasets, built on top of the base model Japanese StableLM-3B-4E1T Base.", "source": "Hugging Face", - "score": 0.00012076742487022366, - "project_name": "bert-base-japanese-v3-marc_ja" + "score": 0.0001099720895325423, + "project_name": "japanese-stablelm-3b-4e1t-instruct" }, { - "url": "https://huggingface.co/vumichien/wav2vec2-large-xlsr-japanese-hiragana", - "downloads": 839, - "description": "Wav2Vec2-Large-XLSR-53-JapaneseFine-tuned facebook/wav2vec2-large-xlsr-53 on Japanese using the Common Voice and Japanese speech corpus of Saruwatari-lab, University of Tokyo JSUT.When using this model, make sure that your speech input is sampled at 16kHz.", + "url": "https://huggingface.co/ken11/albert-base-japanese-v1", + "downloads": 731, + "description": "albert-base-japanese-v1日本語事前学習済みALBERTモデルですHow to useファインチューニングこのモデルはPreTrainedモデルです基本的には各種タスク用にファインチューニングして使用されることを想定していますFill-MaskこのモデルではTokenizerにSentencepieceを利用していますそのままでは[MASK]トークンのあとに余計なトークンが混入する問題があるので、利用する際には以下のようにする必要がありますfor PyTorchfrom transformers import (AlbertForMaskedLM, AlbertTokenizerFast)import torchtokenizer = AlbertTokenizerFast.from_pretrained(\"ken11/albert-base-japanese-v1\")", "source": "Hugging Face", - "score": 0.00011659823874121709, - "project_name": "wav2vec2-large-xlsr-japanese-hiragana" + "score": 0.00010937360197046046, + "project_name": "albert-base-japanese-v1" }, { "url": "https://huggingface.co/TheBloke/japanese-stablelm-instruct-gamma-7B-GGUF", - "downloads": 837, + "downloads": 729, "description": "Chat & support: TheBloke's Discord serverWant to contribute?", "source": "Hugging Face", - "score": 0.00011632029299928332, + "score": 0.00010907435818941952, "project_name": "japanese-stablelm-instruct-gamma-7B-GGUF" }, { - "url": "https://huggingface.co/mmnga/pfnet-nekomata-14b-pfn-qfin-inst-merge-gguf", - "downloads": 828, - "description": "pfnet-nekomata-14b-pfn-qfin-inst-merge-ggufpfnetさんが公開しているnekomata-14b-pfn-qfin-inst-mergeのggufフォーマット変換版です。", + "url": "https://huggingface.co/rinna/japanese-hubert-large", + "downloads": 717, + "description": "rinna/japanese-hubert-largeOverviewThis is a Japanese HuBERT Large model trained by rinna Co.", "source": "Hugging Face", - "score": 0.00011506953716058135, - "project_name": "pfnet-nekomata-14b-pfn-qfin-inst-merge-gguf" + "score": 0.00010727889550317393, + "project_name": "japanese-hubert-large" }, { - "url": "https://huggingface.co/mmnga/tokyotech-llm-Swallow-70b-instruct-v0.1-gguf", - "downloads": 815, - "description": "tokyotech-llm-Swallow-70b-instruct-v0.1-gguftokyotech-llmさんが公開しているSwallow-70b-instruct-v0.1のggufフォーマット変換版です。", + "url": "https://huggingface.co/Aratako/Ninja-v1-RP-expressive-GGUF", + "downloads": 710, + "description": "Ninja-v1-RP-expressive-GGUF概要Aratako/Ninja-v1-RP-expressiveの量子化済みGGUF版です。", "source": "Hugging Face", - "score": 0.00011326288983801183, - "project_name": "tokyotech-llm-Swallow-70b-instruct-v0.1-gguf" + "score": 0.00010623154226953066, + "project_name": "Ninja-v1-RP-expressive-GGUF" }, { - "url": "https://huggingface.co/KoichiYasuoka/bert-base-japanese-wikipedia-ud-head", - "downloads": 813, - "description": "bert-base-japanese-wikipedia-ud-headModel", + "url": "https://huggingface.co/hajime9652/xlnet-japanese", + "downloads": 698, + "description": "XLNet-japaneseModel descriptionThis model require Mecab and senetencepiece with XLNetTokenizer.", "source": "Hugging Face", - "score": 0.00011298494409607806, - "project_name": "bert-base-japanese-wikipedia-ud-head" + "score": 0.00010443607958328509, + "project_name": "xlnet-japanese" }, { - "url": "https://huggingface.co/mmnga/Ninja-v1-128k-gguf", - "downloads": 797, - "description": "Ninja-v1-128k-ggufLocal-Novel-LLM-projectさんが公開しているNinja-v1-128kのggufフォーマット変換版です。", + "url": "https://huggingface.co/mmnga/Mistral-7B-Instruct-v0.3-gguf", + "downloads": 695, + "description": "Mistral-7B-Instruct-v0.3-ggufmistralaiさんが公開しているMistral-7B-Instruct-v0.3のggufフォーマット変換版です。", "source": "Hugging Face", - "score": 0.0001107613781606079, - "project_name": "Ninja-v1-128k-gguf" + "score": 0.00010398721391172368, + "project_name": "Mistral-7B-Instruct-v0.3-gguf" }, { - "url": "https://huggingface.co/mmnga/japanese-stablelm-2-instruct-1_6b-gguf", - "downloads": 795, - "description": "japanese-stablelm-2-instruct-1_6b-ggufstabilityaiさんが公開しているjapanese-stablelm-2-instruct-1_6bのggufフォーマット変換版です。", + "url": "https://huggingface.co/vumichien/wav2vec2-large-xlsr-japanese-hiragana", + "downloads": 690, + "description": "Wav2Vec2-Large-XLSR-53-JapaneseFine-tuned facebook/wav2vec2-large-xlsr-53 on Japanese using the Common Voice and Japanese speech corpus of Saruwatari-lab, University of Tokyo JSUT.When using this model, make sure that your speech input is sampled at 16kHz.", "source": "Hugging Face", - "score": 0.00011048343241867412, - "project_name": "japanese-stablelm-2-instruct-1_6b-gguf" + "score": 0.00010323910445912136, + "project_name": "wav2vec2-large-xlsr-japanese-hiragana" }, { - "url": "https://huggingface.co/mmnga/aixsatoshi-Honyaku-13b-gguf", - "downloads": 790, - "description": "aixsatoshi-Honyaku-13b-ggufaixsatoshiさんが公開しているHonyaku-13bのggufフォーマット変換版です。", + "url": "https://huggingface.co/mmnga/tokyotech-llm-Swallow-MS-7b-instruct-v0.1-gguf", + "downloads": 682, + "description": "tokyotech-llm-Swallow-MS-7b-instruct-v0.1-gguftokyotech-llmさんが公開しているSwallow-MS-7b-instruct-v0.1のggufフォーマット変換版です。", "source": "Hugging Face", - "score": 0.00010978856806383969, - "project_name": "aixsatoshi-Honyaku-13b-gguf" + "score": 0.00010204212933495763, + "project_name": "tokyotech-llm-Swallow-MS-7b-instruct-v0.1-gguf" }, { - "url": "https://huggingface.co/KoichiYasuoka/roberta-small-japanese-luw-upos", - "downloads": 789, - "description": "roberta-small-japanese-luw-uposModel", + "url": "https://huggingface.co/tokyotech-llm/Swallow-70b-instruct-v0.1", + "downloads": 678, + "description": "SwallowOur Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data.", "source": "Hugging Face", - "score": 0.0001096495951928728, - "project_name": "roberta-small-japanese-luw-upos" + "score": 0.00010144364177287577, + "project_name": "Swallow-70b-instruct-v0.1" }, { - "url": "https://huggingface.co/mmnga/aya-23-35B-gguf", - "downloads": 784, - "description": "aya-23-35B-ggufCohereForAIさんが公開しているaya-23-35Bのggufフォーマット変換版です。", + "url": "https://huggingface.co/retrieva-jp/t5-base-long", + "downloads": 677, + "description": "Model card for model IDThis is a T5 v1.1 model, pre-trained on a Japanese corpus.", "source": "Hugging Face", - "score": 0.00010895473083803838, - "project_name": "aya-23-35B-gguf" + "score": 0.0001012940198823553, + "project_name": "t5-base-long" }, { - "url": "https://huggingface.co/mmnga/pfnet-nekomata-14b-pfn-qfin-gguf", - "downloads": 779, - "description": "pfnet-nekomata-14b-pfn-qfin-ggufpfnetさんが公開しているnekomata-14b-pfn-qfinのggufフォーマット変換版です。", + "url": "https://huggingface.co/mmnga/tokyotech-llm-Swallow-70b-instruct-v0.1-gguf", + "downloads": 667, + "description": "tokyotech-llm-Swallow-70b-instruct-v0.1-gguftokyotech-llmさんが公開しているSwallow-70b-instruct-v0.1のggufフォーマット変換版です。", "source": "Hugging Face", - "score": 0.00010825986648320395, - "project_name": "pfnet-nekomata-14b-pfn-qfin-gguf" + "score": 9.979780097715065e-05, + "project_name": "tokyotech-llm-Swallow-70b-instruct-v0.1-gguf" }, { - "url": "https://huggingface.co/mmnga/ELYZA-japanese-CodeLlama-7b-instruct-gguf", - "downloads": 777, - "description": "ELYZA-japanese-CodeLlama-7b-instruct-ggufELYZAさんが公開しているELYZA-japanese-CodeLlama-7b-instructのggufフォーマット変換版です。", + "url": "https://huggingface.co/nk2t/Llama-3-8B-Instruct-japanese-nk2t-v0.3", + "downloads": 662, + "description": "Llama-3-8B-Instruct-JP-nk2t-v0.3Model Details: Built with Meta Llama 3llama-3-8bの日本語継続学習モデルにChatVectorを適用し、さらにQLoraでファインチューニングしたモデルです。", "source": "Hugging Face", - "score": 0.00010798192074127018, - "project_name": "ELYZA-japanese-CodeLlama-7b-instruct-gguf" + "score": 9.904969152454831e-05, + "project_name": "Llama-3-8B-Instruct-japanese-nk2t-v0.3" }, { - "url": "https://huggingface.co/mmnga/alfredplpl-Llama-3-8B-Instruct-Ja-gguf", - "downloads": 774, - "description": "alfredplpl-Llama-3-8B-Instruct-Ja-ggufalfredplplさんが公開しているLlama-3-8B-Instruct-Jaのggufフォーマット変換版です。", + "url": "https://huggingface.co/mmnga/tokyotech-llm-Swallow-7b-instruct-v0.1-gguf", + "downloads": 662, + "description": "tokyotech-llm-Swallow-7b-instruct-v0.1-gguftokyotech-llmさんが公開しているSwallow-7b-instruct-v0.1のggufフォーマット変換版です。", "source": "Hugging Face", - "score": 0.00010756500212836952, - "project_name": "alfredplpl-Llama-3-8B-Instruct-Ja-gguf" + "score": 9.904969152454831e-05, + "project_name": "tokyotech-llm-Swallow-7b-instruct-v0.1-gguf" }, { "url": "https://huggingface.co/mmnga/rinna-japanese-gpt-neox-3.6b-gguf", - "downloads": 771, + "downloads": 648, "description": "rinna/japanese-gpt-neox-3.6brinnaさんが公開しているjapanese-gpt-neox-3.6bのgguf変換版です。", "source": "Hugging Face", - "score": 0.00010714808351546886, + "score": 9.69549850572618e-05, "project_name": "rinna-japanese-gpt-neox-3.6b-gguf" }, { - "url": "https://huggingface.co/nitky/Oumuamua-7b-instruct", - "downloads": 770, - "description": "Oumuamua-7b-instructThis is a merge of pre-trained language models created using mergekit.Output example[INST] <<SYS>>あなたは日本語を話す優秀なアシスタントです。", + "url": "https://huggingface.co/mmnga/ELYZA-japanese-CodeLlama-7b-instruct-gguf", + "downloads": 647, + "description": "ELYZA-japanese-CodeLlama-7b-instruct-ggufELYZAさんが公開しているELYZA-japanese-CodeLlama-7b-instructのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 0.00010700911064450197, - "project_name": "Oumuamua-7b-instruct" + "score": 9.680536316674133e-05, + "project_name": "ELYZA-japanese-CodeLlama-7b-instruct-gguf" }, { "url": "https://huggingface.co/mmnga/line-corp-japanese-large-lm-1.7b-instruction-sft-gguf", - "downloads": 769, + "downloads": 636, "description": "line-corporation/japanese-large-lm-1.7b-instruction-sftline-corporationさんが公開しているjapanese-large-lm-1.7b-instruction-sftのgguf変換版です。", "source": "Hugging Face", - "score": 0.00010687013777353509, + "score": 9.515952237101621e-05, "project_name": "line-corp-japanese-large-lm-1.7b-instruction-sft-gguf" }, { - "url": "https://huggingface.co/mmnga/tokyotech-llm-Swallow-13b-instruct-v0.1-gguf", - "downloads": 755, - "description": "tokyotech-llm-Swallow-13b-instruct-v0.1-gguftokyotech-llmさんが公開しているSwallow-13b-instruct-v0.1のggufフォーマット変換版です。", + "url": "https://huggingface.co/TheBloke/japanese-stablelm-instruct-beta-7B-GGUF", + "downloads": 628, + "description": "Chat & support: TheBloke's Discord serverWant to contribute?", "source": "Hugging Face", - "score": 0.00010492451757999869, - "project_name": "tokyotech-llm-Swallow-13b-instruct-v0.1-gguf" + "score": 9.396254724685248e-05, + "project_name": "japanese-stablelm-instruct-beta-7B-GGUF" }, { - "url": "https://huggingface.co/mmnga/tokyotech-llm-Swallow-7b-instruct-v0.1-gguf", - "downloads": 754, - "description": "tokyotech-llm-Swallow-7b-instruct-v0.1-gguftokyotech-llmさんが公開しているSwallow-7b-instruct-v0.1のggufフォーマット変換版です。", + "url": "https://huggingface.co/mmnga/rinna-japanese-gpt-neox-3.6b-instruction-ppo-gguf", + "downloads": 623, + "description": "rinna/japanese-gpt-neox-3.6b-instruction-pporinnaさんが公開しているjapanese-gpt-neox-3.6b-instruction-ppoのgguf変換版です。", "source": "Hugging Face", - "score": 0.00010478554470903181, - "project_name": "tokyotech-llm-Swallow-7b-instruct-v0.1-gguf" + "score": 9.321443779425014e-05, + "project_name": "rinna-japanese-gpt-neox-3.6b-instruction-ppo-gguf" }, { - "url": "https://huggingface.co/mmnga/YuisekinAIEvol-Mistral-7B-ja-math-v0.1.1-gguf", - "downloads": 752, - "description": "YuisekinAIEvol-Mistral-7B-ja-math-v0.1.1-ggufyuisekiさんが公開しているYuisekinAIEvol-Mistral-7B-ja-math-v0.1.1のggufフォーマット変換版です。", + "url": "https://huggingface.co/TFMC/Japanese-Starling-ChatV-7B", + "downloads": 623, + "description": "Japanese-Starling-ChatV-7Bこのモデルは\"chatntq-ja-7b-v1.0\"をベースにした7Bパラメータの日本語チャットモデルです。", "source": "Hugging Face", - "score": 0.00010450759896709803, - "project_name": "YuisekinAIEvol-Mistral-7B-ja-math-v0.1.1-gguf" + "score": 9.321443779425014e-05, + "project_name": "Japanese-Starling-ChatV-7B" }, { "url": "https://huggingface.co/mmnga/line-corp-japanese-large-lm-1.7b-gguf", - "downloads": 749, + "downloads": 622, "description": "line-corporation/japanese-large-lm-1.7bline-corporationさんが公開しているjapanese-large-lm-1.7bのgguf変換版です。", "source": "Hugging Face", - "score": 0.00010409068035419738, + "score": 9.306481590372969e-05, "project_name": "line-corp-japanese-large-lm-1.7b-gguf" }, { - "url": "https://huggingface.co/mmnga/rinna-japanese-gpt-neox-3.6b-instruction-ppo-gguf", - "downloads": 746, - "description": "rinna/japanese-gpt-neox-3.6b-instruction-pporinnaさんが公開しているjapanese-gpt-neox-3.6b-instruction-ppoのgguf変換版です。", - "source": "Hugging Face", - "score": 0.00010367376174129672, - "project_name": "rinna-japanese-gpt-neox-3.6b-instruction-ppo-gguf" - }, - { - "url": "https://huggingface.co/mmnga/aixsatoshi-Llama-3-8b-Cosmopedia-japanese-gguf", - "downloads": 745, - "description": "aixsatoshi-Llama-3-8b-Cosmopedia-japanese-ggufaixsatoshiさんが公開しているLlama-3-8b-Cosmopedia-japaneseのggufフォーマット変換版です。", + "url": "https://huggingface.co/aken12/splade-japanese-v3", + "downloads": 620, + "description": "Evaluation on MIRACL japaneseThese models don't train on the MIRACL training data.", "source": "Hugging Face", - "score": 0.00010353478887032983, - "project_name": "aixsatoshi-Llama-3-8b-Cosmopedia-japanese-gguf" + "score": 9.276557212268875e-05, + "project_name": "splade-japanese-v3" }, { - "url": "https://huggingface.co/stabilityai/japanese-stablelm-3b-4e1t-instruct", - "downloads": 733, - "description": "Japanese StableLM-3B-4E1T InstructModel DescriptionThis is a 3B-parameter decoder-only Japanese language model fine-tuned on instruction-following datasets, built on top of the base model Japanese StableLM-3B-4E1T Base.", + "url": "https://huggingface.co/mmnga/ELYZA-japanese-Llama-2-13b-fast-gguf", + "downloads": 620, + "description": "ELYZA-japanese-Llama-2-13b-fast-ggufELYZAさんが公開しているELYZA-japanese-Llama-2-13b-fastのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 0.00010186711441872721, - "project_name": "japanese-stablelm-3b-4e1t-instruct" + "score": 9.276557212268875e-05, + "project_name": "ELYZA-japanese-Llama-2-13b-fast-gguf" }, { - "url": "https://huggingface.co/TheBloke/japanese-stablelm-instruct-beta-7B-GGUF", - "downloads": 733, + "url": "https://huggingface.co/TheBloke/japanese-stablelm-instruct-beta-70B-GGUF", + "downloads": 618, "description": "Chat & support: TheBloke's Discord serverWant to contribute?", "source": "Hugging Face", - "score": 0.00010186711441872721, - "project_name": "japanese-stablelm-instruct-beta-7B-GGUF" - }, - { - "url": "https://huggingface.co/abeja/gpt2-large-japanese", - "downloads": 717, - "description": "gpt2-large-japaneseThis repository provides a large sized Japanese GPT-2 model.", - "source": "Hugging Face", - "score": 9.964354848325703e-05, - "project_name": "gpt2-large-japanese" + "score": 9.246632834164782e-05, + "project_name": "japanese-stablelm-instruct-beta-70B-GGUF" }, { - "url": "https://huggingface.co/mmnga/ELYZA-japanese-Llama-2-13b-fast-gguf", - "downloads": 712, - "description": "ELYZA-japanese-Llama-2-13b-fast-ggufELYZAさんが公開しているELYZA-japanese-Llama-2-13b-fastのggufフォーマット変換版です。", + "url": "https://huggingface.co/mmnga/haqishen-Llama-3-8B-Japanese-Instruct-gguf", + "downloads": 614, + "description": "haqishen-Llama-3-8B-Japanese-Instruct-ggufhaqishenさんが公開しているLlama-3-8B-Japanese-Instructのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 9.89486841284226e-05, - "project_name": "ELYZA-japanese-Llama-2-13b-fast-gguf" + "score": 9.186784077956596e-05, + "project_name": "haqishen-Llama-3-8B-Japanese-Instruct-gguf" }, { - "url": "https://huggingface.co/TheBloke/japanese-stablelm-instruct-beta-70B-GGUF", - "downloads": 701, - "description": "Chat & support: TheBloke's Discord serverWant to contribute?", + "url": "https://huggingface.co/mmnga/tokyotech-llm-Swallow-13b-instruct-v0.1-gguf", + "downloads": 609, + "description": "tokyotech-llm-Swallow-13b-instruct-v0.1-gguftokyotech-llmさんが公開しているSwallow-13b-instruct-v0.1のggufフォーマット変換版です。", "source": "Hugging Face", - "score": 9.741998254778686e-05, - "project_name": "japanese-stablelm-instruct-beta-70B-GGUF" + "score": 9.111973132696363e-05, + "project_name": "tokyotech-llm-Swallow-13b-instruct-v0.1-gguf" }, { - "url": "https://huggingface.co/mmnga/aixsatoshi-Ex-karakuri-8x12B-chat-v1-gguf", - "downloads": 691, - "description": "aixsatoshi-Ex-karakuri-8x12B-chat-v1-ggufaixsatoshiさんが公開しているEx-karakuri-8x12B-chat-v1のggufフォーマット変換版です。", + "url": "https://huggingface.co/sbintuitions/tiny-lm-chat", + "downloads": 609, + "description": "tiny-lmThis repository provides a tiny 16M parameters language model for debugging and testing purposes.", "source": "Hugging Face", - "score": 9.6030253838118e-05, - "project_name": "aixsatoshi-Ex-karakuri-8x12B-chat-v1-gguf" + "score": 9.111973132696363e-05, + "project_name": "tiny-lm-chat" }, { - "url": "https://huggingface.co/TFMC/Japanese-Starling-ChatV-7B", - "downloads": 682, - "description": "Japanese-Starling-ChatV-7Bこのモデルは\"chatntq-ja-7b-v1.0\"をベースにした7Bパラメータの日本語チャットモデルです。", + "url": "https://huggingface.co/mmnga/umiyuki-Japanese-Chat-Umievo-itr001-7b-gguf", + "downloads": 595, + "description": "umiyuki-Japanese-Chat-Umievo-itr001-7b-ggufumiyukiさんが公開しているJapanese-Chat-Umievo-itr001-7bのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 9.477949799941603e-05, - "project_name": "Japanese-Starling-ChatV-7B" + "score": 8.902502485967711e-05, + "project_name": "umiyuki-Japanese-Chat-Umievo-itr001-7b-gguf" }, { - "url": "https://huggingface.co/bclavie/fio-base-japanese-v0.1", - "downloads": 676, - "description": "fio-base-japanese-v0.1日本語版は近日公開予定です（日本語を勉強中なので、間違いはご容赦ください！", + "url": "https://huggingface.co/rinna/nue-asr", + "downloads": 594, + "description": "rinna/nue-asrOverview[Paper][GitHub]We propose a novel end-to-end speech recognition model, Nue ASR, which integrates pre-trained speech and language models.", "source": "Hugging Face", - "score": 9.394566077361472e-05, - "project_name": "fio-base-japanese-v0.1" + "score": 8.887540296915664e-05, + "project_name": "nue-asr" }, { - "url": "https://huggingface.co/stabilityai/japanese-stablelm-instruct-alpha-7b-v2", - "downloads": 672, - "description": "Japanese-StableLM-Instruct-Alpha-7B-v2\"A parrot able to speak Japanese, ukiyoe, edo period\" — Stable Diffusion XLModel Descriptionjapanese-stablelm-instruct-alpha-7b-v2 is a 7B parameter decoder-only language models pre-trained built on top of the Japanese-StableLM-Base-Alpha-7B model and further fine-tuned on various instruction-following datasets.", + "url": "https://huggingface.co/mmnga/aya-23-8B-gguf", + "downloads": 594, + "description": "aya-23-8B-ggufCohereForAIさんが公開しているaya-23-8Bのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 9.338976928974718e-05, - "project_name": "japanese-stablelm-instruct-alpha-7b-v2" + "score": 8.887540296915664e-05, + "project_name": "aya-23-8B-gguf" }, { - "url": "https://huggingface.co/ThePioneer/CoolerWaifuDiffusion", - "downloads": 670, - "description": "モデル説明 (model explanation)CoolJapanDiffusion 2.1.1とWaifuDiffusion 1.4 anime epoch2のマージ。", + "url": "https://huggingface.co/minutillamolinara/bert-japanese_finetuned-sentiment-analysis", + "downloads": 593, + "description": "bert-japanese_finetuned-sentiment-analysisThis model was trained from scratch on the Japanese Sentiment Polarity Dictionary dataset.", "source": "Hugging Face", - "score": 9.311182354781341e-05, - "project_name": "CoolerWaifuDiffusion" + "score": 8.872578107863618e-05, + "project_name": "bert-japanese_finetuned-sentiment-analysis" }, { - "url": "https://huggingface.co/jurabi/bert-ner-japanese", - "downloads": 660, - "description": "BERTによる日本語固有表現抽出のモデルBertForTokenClassificationを用いて、日本語の文から固有表現を抽出します。", + "url": "https://huggingface.co/mmnga/lightblue-suzume-llama-3-8B-japanese-gguf", + "downloads": 592, + "description": "lightblue-suzume-llama-3-8B-japanese-gguflightblueさんが公開しているsuzume-llama-3-8B-japaneseのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 9.172209483814455e-05, - "project_name": "bert-ner-japanese" + "score": 8.85761591881157e-05, + "project_name": "lightblue-suzume-llama-3-8B-japanese-gguf" }, { - "url": "https://huggingface.co/studio-ousia/luke-japanese-base-lite", - "downloads": 659, - "description": "luke-japaneseluke-japanese is the Japanese version of LUKE (LanguageUnderstanding with Knowledge-based Embeddings), a pre-trainedknowledge-enhanced contextualized representation of words and entities.", + "url": "https://huggingface.co/mmnga/japanese-stablelm-2-instruct-1_6b-gguf", + "downloads": 590, + "description": "japanese-stablelm-2-instruct-1_6b-ggufstabilityaiさんが公開しているjapanese-stablelm-2-instruct-1_6bのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 9.158312196717767e-05, - "project_name": "luke-japanese-base-lite" + "score": 8.827691540707479e-05, + "project_name": "japanese-stablelm-2-instruct-1_6b-gguf" }, { - "url": "https://huggingface.co/sociocom/MedNERN-CR-JA", - "downloads": 642, - "description": "This is a model for named entity recognition of Japanese medical documents.", + "url": "https://huggingface.co/ku-nlp/deberta-v2-base-japanese-char-wwm", + "downloads": 586, + "description": "Model Card for Japanese character-level DeBERTa V2 baseModel", "source": "Hugging Face", - "score": 8.92205831607406e-05, - "project_name": "MedNERN-CR-JA" + "score": 8.767842784499292e-05, + "project_name": "deberta-v2-base-japanese-char-wwm" }, { - "url": "https://huggingface.co/tokyotech-llm/Swallow-70b-instruct-v0.1", - "downloads": 632, - "description": "SwallowOur Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data.", + "url": "https://huggingface.co/mmnga/c4ai-command-r-plus-gguf", + "downloads": 583, + "description": "c4ai-command-r-plus-ggufCohereForAIさんが公開しているc4ai-command-r-plusのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 8.783085445107175e-05, - "project_name": "Swallow-70b-instruct-v0.1" + "score": 8.722956217343152e-05, + "project_name": "c4ai-command-r-plus-gguf" }, { - "url": "https://huggingface.co/stabilityai/japanese-stablelm-3b-4e1t-base", - "downloads": 630, - "description": "Japanese StableLM-3B-4E1T BaseModel DescriptionThis is a 3B-parameter decoder-only language model with a focus on maximizing Japanese language modeling performance and Japanese downstream task performance.", + "url": "https://huggingface.co/karakuri-ai/karakuri-lm-70b-chat-v0.1", + "downloads": 558, + "description": "KARAKURI LMKARAKURI LM is a pretrained language model that builds upon Llama 2.Our model enhances Llama 2's capabilities by incorporating additional Japanese vocabulary and further pretraining on a mixture of Japanese and multilingual corpora.", "source": "Hugging Face", - "score": 8.755290870913798e-05, - "project_name": "japanese-stablelm-3b-4e1t-base" + "score": 8.348901491041988e-05, + "project_name": "karakuri-lm-70b-chat-v0.1" }, { - "url": "https://huggingface.co/lmg-anon/vntl-llama3-8b-gguf", - "downloads": 630, - "description": "This repository contains some GGUF quantizations of the merge of the VNTL LLaMA 3 8B qlora.", + "url": "https://huggingface.co/mmnga/YuisekinAIEvol-Mistral-7B-ja-math-v0.1.1-gguf", + "downloads": 554, + "description": "YuisekinAIEvol-Mistral-7B-ja-math-v0.1.1-ggufyuisekiさんが公開しているYuisekinAIEvol-Mistral-7B-ja-math-v0.1.1のggufフォーマット変換版です。", "source": "Hugging Face", - "score": 8.755290870913798e-05, - "project_name": "vntl-llama3-8b-gguf" + "score": 8.289052734833801e-05, + "project_name": "YuisekinAIEvol-Mistral-7B-ja-math-v0.1.1-gguf" }, { - "url": "https://huggingface.co/hotchpotch/japanese-reranker-cross-encoder-large-v1", - "downloads": 619, - "description": "hotchpotch/japanese-reranker-cross-encoder-large-v1日本語で学習させた Reranker (CrossEncoder) シリーズです。", + "url": "https://huggingface.co/Local-Novel-LLM-project/Vecteus-v1", + "downloads": 548, + "description": "Our ModelsVecteusNinja-v1Ninja-v1-NSFWNinja-v1-128kNinja-v1-NSFW-128kModel Card for VecTeus-v1.0The Mistral-7B--based Large Language Model (LLM) is an noveldataset fine-tuned version of the Mistral-7B-v0.1VecTeus has the following changes compared to Mistral-7B-v0.1.128k context window (8k context in v0.1)Achieving both high quality Japanese and English generationCan be generated NSFWMemory ability that does not forget even after long-context generationThis model was created with the help of GPUs from the f", "source": "Hugging Face", - "score": 8.602420712850223e-05, - "project_name": "japanese-reranker-cross-encoder-large-v1" + "score": 8.199279600521522e-05, + "project_name": "Vecteus-v1" }, { "url": "https://huggingface.co/Local-Novel-LLM-project/Ocuteus-v1-gguf", - "downloads": 618, + "downloads": 548, "description": "OcuteusのGGUF版です。", "source": "Hugging Face", - "score": 8.588523425753535e-05, + "score": 8.199279600521522e-05, "project_name": "Ocuteus-v1-gguf" }, { - "url": "https://huggingface.co/aken12/splade-japanese-v3", - "downloads": 615, - "description": "Evaluation on MIRACL japaneseThese models don't train on the MIRACL training data.", - "source": "Hugging Face", - "score": 8.54683156446347e-05, - "project_name": "splade-japanese-v3" - }, - { - "url": "https://huggingface.co/Mizuiro-sakura/luke-japanese-base-finetuned-ner", - "downloads": 607, - "description": "このモデルはluke-japanese-baseをファインチューニングして、固有表現抽出（NER）に用いれるようにしたものです。", + "url": "https://huggingface.co/studio-ousia/luke-japanese-base-lite", + "downloads": 547, + "description": "luke-japaneseluke-japanese is the Japanese version of LUKE (LanguageUnderstanding with Knowledge-based Embeddings), a pre-trainedknowledge-enhanced contextualized representation of words and entities.", "source": "Hugging Face", - "score": 8.435653267689961e-05, - "project_name": "luke-japanese-base-finetuned-ner" + "score": 8.184317411469476e-05, + "project_name": "luke-japanese-base-lite" }, { - "url": "https://huggingface.co/stabilityai/japanese-stable-clip-vit-l-16", - "downloads": 598, - "description": "By clicking \"Agree\", you agree to the License Agreement and acknowledge Stability AI's Privacy Policy.", + "url": "https://huggingface.co/stabilityai/japanese-stablelm-3b-4e1t-base", + "downloads": 543, + "description": "Japanese StableLM-3B-4E1T BaseModel DescriptionThis is a 3B-parameter decoder-only language model with a focus on maximizing Japanese language modeling performance and Japanese downstream task performance.", "source": "Hugging Face", - "score": 8.310577683819764e-05, - "project_name": "japanese-stable-clip-vit-l-16" + "score": 8.12446865526129e-05, + "project_name": "japanese-stablelm-3b-4e1t-base" }, { - "url": "https://huggingface.co/second-state/Llama-3-8B-Japanese-Instruct-GGUF", - "downloads": 590, - "description": "Llama-3-8B-Japanese-Instruct-GGUFOriginal Modelhaqishen/Llama-3-8B-Japanese-InstructRun with LlamaEdgeLlamaEdge version: v0.10.1 and abovePrompt templatePrompt type: llama-3-chatPrompt string<|begin_of_text|><|start_header_id|>system<|end_header_id|>{{ system_prompt }}<|eot_id|><|start_header_id|>user<|end_header_id|>", + "url": "https://huggingface.co/llm-book/bert-base-japanese-v3-crf-ner-wikipedia-dataset", + "downloads": 530, + "description": "llm-book/bert-base-japanese-v3-crf-ner-wikipedia-dataset「大規模言語モデル入門」の第6章で紹介している固有表現認識のモデルです。", "source": "Hugging Face", - "score": 8.199399387046256e-05, - "project_name": "Llama-3-8B-Japanese-Instruct-GGUF" + "score": 7.929960197584684e-05, + "project_name": "bert-base-japanese-v3-crf-ner-wikipedia-dataset" }, { "url": "https://huggingface.co/sazyou-roukaku/LittleStepMix", - "downloads": 582, + "downloads": 526, "description": "License:CreativeML Open RAIL-MAdditional Copyright: sazyou_roukaku (TwitterID @sazyou_roukaku) as of June 25, 2023このモデルは『CreativeML Open RAIL-M』でLicenseそのものに変更はありません。", "source": "Hugging Face", - "score": 8.088221090272747e-05, + "score": 7.870111441376498e-05, "project_name": "LittleStepMix" }, { "url": "https://huggingface.co/hotchpotch/japanese-reranker-cross-encoder-base-v1", - "downloads": 547, + "downloads": 520, "description": "hotchpotch/japanese-reranker-cross-encoder-base-v1日本語で学習させた Reranker (CrossEncoder) シリーズです。", "source": "Hugging Face", - "score": 7.601816041888647e-05, + "score": 7.780338307064218e-05, "project_name": "japanese-reranker-cross-encoder-base-v1" }, - { - "url": "https://huggingface.co/maddes8cht/stabilityai-japanese-stablelm-3b-4e1t-instruct-gguf", - "downloads": 528, - "description": "I'm constantly enhancing these model descriptions to provide you with the most relevant and comprehensive informationjapanese-stablelm-3b-4e1t-instruct - GGUFModel creator: stabilityaiOriginal model: japanese-stablelm-3b-4e1t-instructStableLMThis is a Model based on StableLM.Stablelm is a familiy of Language Models by Stability AI.Note:Current (as of 2023-11-15) implementations of Llama.cpp only support GPU offloading up to 34 Layers with these StableLM Models.", - "source": "Hugging Face", - "score": 7.337767587051564e-05, - "project_name": "stabilityai-japanese-stablelm-3b-4e1t-instruct-gguf" - }, - { - "url": "https://huggingface.co/mmnga/Qwen1.5-110B-Chat-gguf", - "downloads": 528, - "description": "Qwen1.5-110B-Chat-ggufQwenさんが公開しているQwen1.5-110B-Chatのggufフォーマット変換版です。", - "source": "Hugging Face", - "score": 7.337767587051564e-05, - "project_name": "Qwen1.5-110B-Chat-gguf" - }, { "url": "https://huggingface.co/rinna/japanese-wav2vec2-base", - "downloads": 519, + "downloads": 520, "description": "rinna/japanese-wav2vec2-baseOverviewThis is a Japanese wav2vec 2.0 Base model trained by rinna Co.", "source": "Hugging Face", - "score": 7.212692003181367e-05, + "score": 7.780338307064218e-05, "project_name": "japanese-wav2vec2-base" }, { - "url": "https://huggingface.co/classla/xlm-roberta-base-multilingual-text-genre-classifier", - "downloads": 515, - "description": "X-GENRE classifier - multilingual text genre classifierText classification model based on xlm-roberta-base and fine-tuned on a combination of three genre datasets: Slovene GINCO dataset (Kuzman et al.", + "url": "https://huggingface.co/elyza/Llama-3-ELYZA-JP-8B-GGUF", + "downloads": 519, + "description": "Llama-3-ELYZA-JP-8B-GGUFModel DescriptionLlama-3-ELYZA-JP-8B is a large language model trained by ELYZA, Inc.Based on meta-llama/Meta-Llama-3-8B-Instruct, it has been enhanced for Japanese usage through additional pre-training and instruction tuning.", "source": "Hugging Face", - "score": 7.157102854794612e-05, - "project_name": "xlm-roberta-base-multilingual-text-genre-classifier" + "score": 7.765376118012171e-05, + "project_name": "Llama-3-ELYZA-JP-8B-GGUF" }, { - "url": "https://huggingface.co/ku-nlp/deberta-v2-base-japanese-char-wwm", - "downloads": 515, - "description": "Model Card for Japanese character-level DeBERTa V2 baseModel", + "url": "https://huggingface.co/ThePioneer/CoolerWaifuDiffusion", + "downloads": 518, + "description": "モデル説明 (model explanation)CoolJapanDiffusion 2.1.1とWaifuDiffusion 1.4 anime epoch2のマージ。", "source": "Hugging Face", - "score": 7.157102854794612e-05, - "project_name": "deberta-v2-base-japanese-char-wwm" + "score": 7.750413928960125e-05, + "project_name": "CoolerWaifuDiffusion" }, { - "url": "https://huggingface.co/TareHimself/manga-ocr-base", - "downloads": 511, - "description": "Original ModelOptical character recognition for Japanese text, with the main focus being Japanese manga.", + "url": "https://huggingface.co/mmnga/Ninja-v1-128k-gguf", + "downloads": 509, + "description": "Ninja-v1-128k-ggufLocal-Novel-LLM-projectさんが公開しているNinja-v1-128kのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 7.101513706407859e-05, - "project_name": "manga-ocr-base" + "score": 7.615754227491705e-05, + "project_name": "Ninja-v1-128k-gguf" }, { - "url": "https://huggingface.co/rinna/japanese-gpt-neox-small", - "downloads": 509, - "description": "japanese-gpt-neox-smallThis repository provides a small-sized Japanese GPT-NeoX model.", + "url": "https://huggingface.co/mmnga/pfnet-nekomata-14b-pfn-qfin-gguf", + "downloads": 502, + "description": "pfnet-nekomata-14b-pfn-qfin-ggufpfnetさんが公開しているnekomata-14b-pfn-qfinのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 7.073719132214481e-05, - "project_name": "japanese-gpt-neox-small" + "score": 7.51101890412738e-05, + "project_name": "pfnet-nekomata-14b-pfn-qfin-gguf" }, { - "url": "https://huggingface.co/mmnga/ELYZA-japanese-CodeLlama-7b-gguf", - "downloads": 495, - "description": "ELYZA-japanese-CodeLlama-7b-ggufELYZAさんが公開しているELYZA-japanese-CodeLlama-7b-instructのggufフォーマット変換版です。", + "url": "https://huggingface.co/cyberagent/open-calm-medium", + "downloads": 502, + "description": "OpenCALM-MediumModel DescriptionOpenCALM is a suite of decoder-only language models pre-trained on Japanese datasets, developed by", "source": "Hugging Face", - "score": 6.879157112860841e-05, - "project_name": "ELYZA-japanese-CodeLlama-7b-gguf" + "score": 7.51101890412738e-05, + "project_name": "open-calm-medium" }, { - "url": "https://huggingface.co/second-state/ELYZA-japanese-Llama-2-13b-fast-instruct-GGUF", - "downloads": 493, - "description": "ELYZA-japanese-Llama-2-13b-fast-instruct-GGUFOriginal Modelelyza/ELYZA-japanese-Llama-2-13b-fast-instructRun with LlamaEdgeLlamaEdge version: v0.2.8 and abovePrompt templatePrompt type: llama-2-chatPrompt string<s>[INST] <<SYS>>{{ system_prompt }}<</SYS>>{{ user_msg_1 }}", + "url": "https://huggingface.co/sonoisa/t5-base-japanese-title-generation", + "downloads": 502, + "description": "記事本文からタイトルを生成するモデルSEE: https://qiita.com/sonoisa/items/a9af64ff641f0bbfed44", "source": "Hugging Face", - "score": 6.851362538667464e-05, - "project_name": "ELYZA-japanese-Llama-2-13b-fast-instruct-GGUF" + "score": 7.51101890412738e-05, + "project_name": "t5-base-japanese-title-generation" }, { - "url": "https://huggingface.co/ku-nlp/bart-base-japanese", - "downloads": 489, - "description": "Model Card for Japanese BART baseModel", + "url": "https://huggingface.co/mmnga/aixsatoshi-Llama-3-8b-Cosmopedia-japanese-gguf", + "downloads": 499, + "description": "aixsatoshi-Llama-3-8b-Cosmopedia-japanese-ggufaixsatoshiさんが公開しているLlama-3-8b-Cosmopedia-japaneseのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 6.79577339028071e-05, - "project_name": "bart-base-japanese" + "score": 7.46613233697124e-05, + "project_name": "aixsatoshi-Llama-3-8b-Cosmopedia-japanese-gguf" }, { - "url": "https://huggingface.co/gaianet/Llama-3-8B-Japanese-Instruct-GGUF", - "downloads": 478, - "description": "Llama-3-8B-Japanese-Instruct-GGUFOriginal Modelhaqishen/Llama-3-8B-Japanese-InstructRun with GaianetPrompt template:prompt template: llama-3-chatContext size:chat_ctx_size: 4096Run with GaiaNet:Quick start: https://docs.gaianet.ai/node-guide/quick-startCustomize your node: https://docs.gaianet.ai/node-guide/customizeQuantized GGUF ModelsNameQuant methodBitsSizeUse caseLlama-3-8B-Japanese-Instruct-Q2_K.ggufQ2_K23.18 GBsmallest, significant quality loss - not recommended for most purposesLlama-3-8B-Japanese-I", + "url": "https://huggingface.co/hotchpotch/japanese-reranker-cross-encoder-large-v1", + "downloads": 497, + "description": "hotchpotch/japanese-reranker-cross-encoder-large-v1日本語で学習させた Reranker (CrossEncoder) シリーズです。", "source": "Hugging Face", - "score": 6.642903232217136e-05, - "project_name": "Llama-3-8B-Japanese-Instruct-GGUF" + "score": 7.436207958867147e-05, + "project_name": "japanese-reranker-cross-encoder-large-v1" }, { - "url": "https://huggingface.co/nlp-waseda/roberta-large-japanese-seq512", - "downloads": 476, - "description": "nlp-waseda/roberta-large-japanese-seq512Model descriptionThis is a Japanese RoBERTa large model pretrained on Japanese Wikipedia and the Japanese portion of CC-100 with the maximum sequence length of 512.How to useYou can use this model for masked language modeling as follows:from transformers import AutoTokenizer, AutoModelForMaskedLMtokenizer = AutoTokenizer.from_pretrained(\"nlp-waseda/roberta-large-japanese-seq512\")", + "url": "https://huggingface.co/TareHimself/manga-ocr-base", + "downloads": 497, + "description": "Original ModelOptical character recognition for Japanese text, with the main focus being Japanese manga.", "source": "Hugging Face", - "score": 6.615108658023758e-05, - "project_name": "roberta-large-japanese-seq512" + "score": 7.436207958867147e-05, + "project_name": "manga-ocr-base" }, { - "url": "https://huggingface.co/Aratako/Ninja-v1-RP-GGUF", - "downloads": 471, - "description": "Ninja-v1-RP-GGUF概要Aratako/Ninja-v1-RPの量子化済みGGUF版です。", + "url": "https://huggingface.co/mmnga/Meta-Llama-3-8B-Instruct-gguf", + "downloads": 495, + "description": "Meta-Llama-3-8B-Instruct-ggufmeta-llamaさんが公開しているMeta-Llama-3-8B-Instructのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 6.545622222540315e-05, - "project_name": "Ninja-v1-RP-GGUF" + "score": 7.406283580763054e-05, + "project_name": "Meta-Llama-3-8B-Instruct-gguf" }, { - "url": "https://huggingface.co/cyberagent/open-calm-medium", - "downloads": 470, - "description": "OpenCALM-MediumModel DescriptionOpenCALM is a suite of decoder-only language models pre-trained on Japanese datasets, developed by", + "url": "https://huggingface.co/mmnga/DataPilot-ArrowPro-7B-RobinHood-gguf", + "downloads": 487, + "description": "DataPilot-ArrowPro-7B-RobinHood-ggufDataPilotさんが公開しているArrowPro-7B-RobinHoodのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 6.531724935443627e-05, - "project_name": "open-calm-medium" + "score": 7.286586068346681e-05, + "project_name": "DataPilot-ArrowPro-7B-RobinHood-gguf" }, { "url": "https://huggingface.co/cyberagent/xlm-roberta-large-jnli-jsick", - "downloads": 467, + "downloads": 476, "description": "Japanese Natural Language Inference ModelThis model was trained using SentenceTransformers Cross-Encoder class, gradient accumulation PR, and the code from CyberAgentAILab/japanese-nli-model.", "source": "Hugging Face", - "score": 6.490033074153562e-05, + "score": 7.12200198877417e-05, "project_name": "xlm-roberta-large-jnli-jsick" }, { - "url": "https://huggingface.co/maddes8cht/stabilityai-japanese-stablelm-3b-4e1t-base-gguf", - "downloads": 465, - "description": "I'm constantly enhancing these model descriptions to provide you with the most relevant and comprehensive informationjapanese-stablelm-3b-4e1t-base - GGUFModel creator: stabilityaiOriginal model: japanese-stablelm-3b-4e1t-baseStableLMThis is a Model based on StableLM.Stablelm is a familiy of Language Models by Stability AI.Note:Current (as of 2023-11-15) implementations of Llama.cpp only support GPU offloading up to 34 Layers with these StableLM Models.", + "url": "https://huggingface.co/mmnga/ryota39-Phi-3-mini-4k-instruct-dpo-gguf", + "downloads": 475, + "description": "ryota39-Phi-3-mini-4k-instruct-dpo-ggufryota39さんが公開しているPhi-3-mini-4k-instruct-dpoのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 6.462238499960184e-05, - "project_name": "stabilityai-japanese-stablelm-3b-4e1t-base-gguf" + "score": 7.107039799722122e-05, + "project_name": "ryota39-Phi-3-mini-4k-instruct-dpo-gguf" + }, + { + "url": "https://huggingface.co/mmnga/pfnet-nekomata-14b-pfn-qfin-inst-merge-gguf", + "downloads": 469, + "description": "pfnet-nekomata-14b-pfn-qfin-inst-merge-ggufpfnetさんが公開しているnekomata-14b-pfn-qfin-inst-mergeのggufフォーマット変換版です。", + "source": "Hugging Face", + "score": 7.017266665409843e-05, + "project_name": "pfnet-nekomata-14b-pfn-qfin-inst-merge-gguf" }, { "url": "https://huggingface.co/cameltech/japanese-gpt-1b-PII-masking", - "downloads": 462, + "downloads": 469, "description": "japanese-gpt-1b-PII-maskingModel Descriptionjapanese-gpt-1b-PII-masking は、 日本語事前学習済み1B GPTモデルをベースとして、日本語の文章から個人情報をマスキングするように学習したモデルです。", "source": "Hugging Face", - "score": 6.420546638670119e-05, + "score": 7.017266665409843e-05, "project_name": "japanese-gpt-1b-PII-masking" }, { - "url": "https://huggingface.co/tohoku-nlp/bert-large-japanese-char-v2", - "downloads": 459, - "description": "BERT large Japanese (character-level tokenization with whole word masking, CC-100 and jawiki-20230102)This is a BERT model pretrained on texts in the Japanese language.", + "url": "https://huggingface.co/nlp-waseda/roberta-large-japanese-seq512", + "downloads": 467, + "description": "nlp-waseda/roberta-large-japanese-seq512Model descriptionThis is a Japanese RoBERTa large model pretrained on Japanese Wikipedia and the Japanese portion of CC-100 with the maximum sequence length of 512.How to useYou can use this model for masked language modeling as follows:from transformers import AutoTokenizer, AutoModelForMaskedLMtokenizer = AutoTokenizer.from_pretrained(\"nlp-waseda/roberta-large-japanese-seq512\")", "source": "Hugging Face", - "score": 6.378854777380053e-05, - "project_name": "bert-large-japanese-char-v2" + "score": 6.98734228730575e-05, + "project_name": "roberta-large-japanese-seq512" }, { - "url": "https://huggingface.co/mmnga/aya-23-8B-gguf", - "downloads": 458, - "description": "aya-23-8B-ggufCohereForAIさんが公開しているaya-23-8Bのggufフォーマット変換版です。", + "url": "https://huggingface.co/second-state/Llama-3-8B-Japanese-Instruct-GGUF", + "downloads": 467, + "description": "Llama-3-8B-Japanese-Instruct-GGUFOriginal Modelhaqishen/Llama-3-8B-Japanese-InstructRun with LlamaEdgeLlamaEdge version: v0.10.1 and abovePrompt templatePrompt type: llama-3-chatPrompt string<|begin_of_text|><|start_header_id|>system<|end_header_id|>{{ system_prompt }}<|eot_id|><|start_header_id|>user<|end_header_id|>", "source": "Hugging Face", - "score": 6.364957490283364e-05, - "project_name": "aya-23-8B-gguf" + "score": 6.98734228730575e-05, + "project_name": "Llama-3-8B-Japanese-Instruct-GGUF" }, { - "url": "https://huggingface.co/llm-book/t5-base-long-livedoor-news-corpus", - "downloads": 455, - "description": "llm-book/t5-base-long-livedoor-news-corpus「大規模言語モデル入門」の第7章で紹介している要約生成のモデルです。", + "url": "https://huggingface.co/aipib/karasu-lora-jp-qa-chat", + "downloads": 465, + "description": "karasu-lora-jp-qa-chatkarasu fine tuned model by lora method with the original Q&A dataset.", "source": "Hugging Face", - "score": 6.323265628993298e-05, - "project_name": "t5-base-long-livedoor-news-corpus" + "score": 6.957417909201656e-05, + "project_name": "karasu-lora-jp-qa-chat" }, { - "url": "https://huggingface.co/stabilityai/japanese-stable-diffusion-xl", - "downloads": 449, - "description": "By clicking \"Agree\", you agree to the License Agreement and acknowledge Stability AI's Privacy Policy.", + "url": "https://huggingface.co/ku-nlp/bart-base-japanese", + "downloads": 459, + "description": "Model Card for Japanese BART baseModel", "source": "Hugging Face", - "score": 6.239881906413167e-05, - "project_name": "japanese-stable-diffusion-xl" + "score": 6.867644774889377e-05, + "project_name": "bart-base-japanese" }, { "url": "https://huggingface.co/llm-book/bert-base-japanese-v3-jnli", - "downloads": 449, + "downloads": 457, "description": "bert-base-japanese-v3-jnli「大規模言語モデル入門」の第5章で紹介している(自然言語推論)のモデルです。", "source": "Hugging Face", - "score": 6.239881906413167e-05, + "score": 6.837720396785283e-05, "project_name": "bert-base-japanese-v3-jnli" }, { - "url": "https://huggingface.co/owner203/japanese-llama-2-7b-gguf", - "downloads": 448, - "description": "Japanese-LLaMA-2-7B-GGUFJapanese-LLaMA-2-7B-GGUFはJapanese-LLaMA-2-7BのGGUF形式です。", - "source": "Hugging Face", - "score": 6.225984619316479e-05, - "project_name": "japanese-llama-2-7b-gguf" - }, - { - "url": "https://huggingface.co/sonoisa/t5-base-japanese-title-generation", - "downloads": 447, - "description": "記事本文からタイトルを生成するモデルSEE: https://qiita.com/sonoisa/items/a9af64ff641f0bbfed44", + "url": "https://huggingface.co/ku-nlp/deberta-v2-large-japanese", + "downloads": 455, + "description": "Model Card for Japanese DeBERTa V2 largeModel descriptionThis is a Japanese DeBERTa V2 large model pre-trained on Japanese Wikipedia, the Japanese portion of CC-100, and theJapanese portion of OSCAR.How to useYou can use this model for masked language modeling as follows:from transformers import AutoTokenizer, AutoModelForMaskedLMtokenizer = AutoTokenizer.from_pretrained('ku-nlp/deberta-v2-large-japanese')", "source": "Hugging Face", - "score": 6.21208733221979e-05, - "project_name": "t5-base-japanese-title-generation" + "score": 6.80779601868119e-05, + "project_name": "deberta-v2-large-japanese" }, { - "url": "https://huggingface.co/retrieva-jp/t5-large-long", - "downloads": 442, - "description": "Model card for model IDThis is a T5 v1.1 model, pre-trained on a Japanese corpus.", + "url": "https://huggingface.co/owner203/japanese-llama-2-7b-gguf", + "downloads": 451, + "description": "Japanese-LLaMA-2-7B-GGUFJapanese-LLaMA-2-7B-GGUFはJapanese-LLaMA-2-7BのGGUF形式です。", "source": "Hugging Face", - "score": 6.142600896736348e-05, - "project_name": "t5-large-long" + "score": 6.747947262473004e-05, + "project_name": "japanese-llama-2-7b-gguf" }, { - "url": "https://huggingface.co/sbintuitions/tiny-lm-chat", - "downloads": 439, - "description": "tiny-lmThis repository provides a tiny 16M parameters language model for debugging and testing purposes.", + "url": "https://huggingface.co/OrionStarAI/Orion-14B-Chat-Int4", + "downloads": 449, + "description": "Orion-14B🌐English | 🇨🇳中文 | 🇯🇵日本語 | 🇰🇷한국어🤗", "source": "Hugging Face", - "score": 6.1009090354462815e-05, - "project_name": "tiny-lm-chat" + "score": 6.718022884368912e-05, + "project_name": "Orion-14B-Chat-Int4" }, { - "url": "https://huggingface.co/stockmark/gpt-neox-japanese-1.4b", - "downloads": 439, - "description": "stockmark/gpt-neox-japanese-1.4bThis repository provides a GPT-NeoX based model with 1.4B parameters pre-trained on Japanese corpus of about 20B tokens.", + "url": "https://huggingface.co/rinna/nekomata-7b", + "downloads": 443, + "description": "rinna/nekomata-7bOverviewWe conduct continual pre-training of qwen-7b on 30B tokens from a mixture of Japanese and English datasets.", "source": "Hugging Face", - "score": 6.1009090354462815e-05, - "project_name": "gpt-neox-japanese-1.4b" + "score": 6.628249750056632e-05, + "project_name": "nekomata-7b" }, { - "url": "https://huggingface.co/aipib/karasu-lora-jp-qa-chat", - "downloads": 438, - "description": "karasu-lora-jp-qa-chatkarasu fine tuned model by lora method with the original Q&A dataset.", + "url": "https://huggingface.co/llm-book/t5-base-long-livedoor-news-corpus", + "downloads": 441, + "description": "llm-book/t5-base-long-livedoor-news-corpus「大規模言語モデル入門」の第7章で紹介している要約生成のモデルです。", "source": "Hugging Face", - "score": 6.087011748349593e-05, - "project_name": "karasu-lora-jp-qa-chat" + "score": 6.598325371952539e-05, + "project_name": "t5-base-long-livedoor-news-corpus" }, { - "url": "https://huggingface.co/MCZK/Assistance-7B-GGUF", + "url": "https://huggingface.co/rinna/japanese-gpt-neox-small", "downloads": 436, - "description": "Local-Novel-LLM-project様の Assistance をGGUF形式に変換したものです。", - "source": "Hugging Face", - "score": 6.059217174156216e-05, - "project_name": "Assistance-7B-GGUF" - }, - { - "url": "https://huggingface.co/NTQAI/wav2vec2-large-japanese", - "downloads": 425, - "description": "Wav2Vec2-Large-JapaneseFine-tuned facebook/wav2vec2-large-xlsr-53 on Japanese using the Common Voice, JSUT, TEDxJP and some other data.", - "source": "Hugging Face", - "score": 5.906347016092641e-05, - "project_name": "wav2vec2-large-japanese" - }, - { - "url": "https://huggingface.co/tsmatz/mt5_summarize_japanese", - "downloads": 424, - "description": "mt5_summarize_japanese(Japanese caption : 日本語の要約のモデル)This model is a fine-tuned version of google/mt5-small trained for Japanese summarization.", + "description": "japanese-gpt-neox-smallThis repository provides a small-sized Japanese GPT-NeoX model.", "source": "Hugging Face", - "score": 5.892449728995953e-05, - "project_name": "mt5_summarize_japanese" + "score": 6.523514426692306e-05, + "project_name": "japanese-gpt-neox-small" }, { - "url": "https://huggingface.co/retrieva-jp/t5-base-long", - "downloads": 413, - "description": "Model card for model IDThis is a T5 v1.1 model, pre-trained on a Japanese corpus.", + "url": "https://huggingface.co/mmnga/alfredplpl-Llama-3-8B-Instruct-Ja-gguf", + "downloads": 428, + "description": "alfredplpl-Llama-3-8B-Instruct-Ja-ggufalfredplplさんが公開しているLlama-3-8B-Instruct-Jaのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 5.7395795709323784e-05, - "project_name": "t5-base-long" + "score": 6.403816914275933e-05, + "project_name": "alfredplpl-Llama-3-8B-Instruct-Ja-gguf" }, { - "url": "https://huggingface.co/Fugaku-LLM/Fugaku-LLM-13B-instruct-gguf", - "downloads": 408, - "description": "Fugaku-LLM利用規約この利用規約（以下「本規約」といいます）は、富士通株式会社、国立研究開発法人理化学研究所、国立大学法人東京工業大学、国立大学法人東北大学、株式会社サイバーエージェント、国立大学法人東海国立大学機構、及び株式会社Kotoba Technologies Japan (以下「開発者」といいます)による、スーパーコンピュータ「富岳」政策対応枠における大規模言語モデル分散並列学習手法の開発の成果物として公開する大規模言語モデル（以下「Fugaku-LLM」といいます）の利用に関する条件を定めるものです。", + "url": "https://huggingface.co/sonoisa/clip-vit-b-32-japanese-v1", + "downloads": 427, + "description": "日本語版CLIPモデルThis is a CLIP text/image encoder model for Japanese.", "source": "Hugging Face", - "score": 5.6700931354489356e-05, - "project_name": "Fugaku-LLM-13B-instruct-gguf" + "score": 6.388854725223887e-05, + "project_name": "clip-vit-b-32-japanese-v1" }, { - "url": "https://huggingface.co/retrieva-jp/t5-small-short", - "downloads": 407, - "description": "Model card for model IDThis is a T5 v1.1 model, pre-trained on a Japanese corpus.", + "url": "https://huggingface.co/izumi-lab/deberta-v2-base-japanese", + "downloads": 421, + "description": "DeBERTa V2 base JapaneseThis is a DeBERTaV2 model pretrained on Japanese texts.", "source": "Hugging Face", - "score": 5.656195848352247e-05, - "project_name": "t5-small-short" + "score": 6.299081590911607e-05, + "project_name": "deberta-v2-base-japanese" }, { - "url": "https://huggingface.co/NTQAI/chatntq-ja-7b-v1.0", - "downloads": 406, - "description": "ChatNTQ JA 7B V1.0Model", + "url": "https://huggingface.co/Aratako/Ninja-v1-RP-expressive-breadcrumbs", + "downloads": 418, + "description": "Ninja-v1-RP-expressive-breadcrumbsGGUF版はこちら/Click here for the GGUF version概要This is a merge of pre-trained language models created using mergekit.", "source": "Hugging Face", - "score": 5.642298561255559e-05, - "project_name": "chatntq-ja-7b-v1.0" + "score": 6.254195023755468e-05, + "project_name": "Ninja-v1-RP-expressive-breadcrumbs" }, { - "url": "https://huggingface.co/mmnga/SakanaAI-EvoLLM-JP-A-v1-7B-gguf", - "downloads": 406, - "description": "SakanaAI-EvoLLM-JP-A-v1-7B-ggufSakanaAIさんが公開しているEvoLLM-JP-A-v1-7Bのggufフォーマット変換版です。", + "url": "https://huggingface.co/second-state/ELYZA-japanese-Llama-2-13b-fast-instruct-GGUF", + "downloads": 408, + "description": "ELYZA-japanese-Llama-2-13b-fast-instruct-GGUFOriginal Modelelyza/ELYZA-japanese-Llama-2-13b-fast-instructRun with LlamaEdgeLlamaEdge version: v0.2.8 and abovePrompt templatePrompt type: llama-2-chatPrompt string<s>[INST] <<SYS>>{{ system_prompt }}<</SYS>>{{ user_msg_1 }}", "source": "Hugging Face", - "score": 5.642298561255559e-05, - "project_name": "SakanaAI-EvoLLM-JP-A-v1-7B-gguf" + "score": 6.104573133235002e-05, + "project_name": "ELYZA-japanese-Llama-2-13b-fast-instruct-GGUF" }, { - "url": "https://huggingface.co/Aratako/Ninja-v1-RP-expressive-v2", - "downloads": 398, - "description": "Ninja-v1-RP-expressive-v2GGUF版はこちら/Click here for the GGUF version概要This is a merge of pre-trained language models created using mergekit.Aratako/Ninja-v1-RP-expressiveと同じコンセプトで、ライセンスがCC-BY-NCのものがマージ元に含まれないようにレシピを変更したモデルです。", + "url": "https://huggingface.co/NTQAI/wav2vec2-large-japanese", + "downloads": 399, + "description": "Wav2Vec2-Large-JapaneseFine-tuned facebook/wav2vec2-large-xlsr-53 on Japanese using the Common Voice, JSUT, TEDxJP and some other data.", "source": "Hugging Face", - "score": 5.53112026448205e-05, - "project_name": "Ninja-v1-RP-expressive-v2" + "score": 5.969913431766583e-05, + "project_name": "wav2vec2-large-japanese" }, { - "url": "https://huggingface.co/Aratako/c4ai-command-r-v01-japanese-instruct-GGUF", + "url": "https://huggingface.co/inu-ai/dolly-japanese-gpt-1b", "downloads": 396, - "description": "c4ai-command-r-v01-japanese-instruct-GGUF概要Aratako/c4ai-command-r-v01-japanese-instructの量子化済みGGUF版です。", + "description": "更新履歴2023年5月7日「oasst1-89k-ja」データセットを追加して対話システムに対応しました。", "source": "Hugging Face", - "score": 5.5033256902886734e-05, - "project_name": "c4ai-command-r-v01-japanese-instruct-GGUF" + "score": 5.925026864610443e-05, + "project_name": "dolly-japanese-gpt-1b" }, { - "url": "https://huggingface.co/minutillamolinara/bert-japanese_finetuned-sentiment-analysis", + "url": "https://huggingface.co/stabilityai/japanese-stable-clip-vit-l-16", "downloads": 394, - "description": "bert-japanese_finetuned-sentiment-analysisThis model was trained from scratch on the Japanese Sentiment Polarity Dictionary dataset.", - "source": "Hugging Face", - "score": 5.475531116095296e-05, - "project_name": "bert-japanese_finetuned-sentiment-analysis" - }, - { - "url": "https://huggingface.co/rinna/nekomata-7b", - "downloads": 393, - "description": "rinna/nekomata-7bOverviewWe conduct continual pre-training of qwen-7b on 30B tokens from a mixture of Japanese and English datasets.", + "description": "By clicking \"Agree\", you agree to the License Agreement and acknowledge Stability AI's Privacy Policy.", "source": "Hugging Face", - "score": 5.461633828998607e-05, - "project_name": "nekomata-7b" + "score": 5.89510248650635e-05, + "project_name": "japanese-stable-clip-vit-l-16" }, { "url": "https://huggingface.co/sociocom/MedNER-CR-JA", - "downloads": 389, + "downloads": 393, "description": "This is a model for named entity recognition of Japanese medical documents.", "source": "Hugging Face", - "score": 5.406044680611853e-05, + "score": 5.880140297454303e-05, "project_name": "MedNER-CR-JA" }, { - "url": "https://huggingface.co/sonoisa/clip-vit-b-32-japanese-v1", - "downloads": 387, - "description": "日本語版CLIPモデルThis is a CLIP text/image encoder model for Japanese.", - "source": "Hugging Face", - "score": 5.378250106418476e-05, - "project_name": "clip-vit-b-32-japanese-v1" - }, - { - "url": "https://huggingface.co/nk2t/Llama-3-8B-Instruct-japanese-nk2t-v0.3", - "downloads": 383, - "description": "Llama-3-8B-Instruct-JP-nk2t-v0.3Model Details: Built with Meta Llama 3llama-3-8bの日本語継続学習モデルにChatVectorを適用し、さらにQLoraでファインチューニングしたモデルです。", + "url": "https://huggingface.co/mmnga/aixsatoshi-Ex-karakuri-8x12B-chat-v1-gguf", + "downloads": 391, + "description": "aixsatoshi-Ex-karakuri-8x12B-chat-v1-ggufaixsatoshiさんが公開しているEx-karakuri-8x12B-chat-v1のggufフォーマット変換版です。", "source": "Hugging Face", - "score": 5.3226609580317215e-05, - "project_name": "Llama-3-8B-Instruct-japanese-nk2t-v0.3" + "score": 5.85021591935021e-05, + "project_name": "aixsatoshi-Ex-karakuri-8x12B-chat-v1-gguf" }, { - "url": "https://huggingface.co/llm-book/bert-base-japanese-v3-jsts", - "downloads": 369, - "description": "bert-base-japanese-v3-jsts「大規模言語モデル入門」の第5章で紹介している(意味類似度計算)のモデルです。", + "url": "https://huggingface.co/NTQAI/chatntq-ja-7b-v1.0", + "downloads": 391, + "description": "ChatNTQ JA 7B V1.0Model", "source": "Hugging Face", - "score": 5.128098938678082e-05, - "project_name": "bert-base-japanese-v3-jsts" + "score": 5.85021591935021e-05, + "project_name": "chatntq-ja-7b-v1.0" }, { - "url": "https://huggingface.co/studio-ousia/luke-japanese-base", - "downloads": 366, - "description": "luke-japaneseluke-japanese is the Japanese version of LUKE (Language Understanding with Knowledge-based Embeddings), a pre-trained knowledge-enhanced contextualized representation of words and entities.", + "url": "https://huggingface.co/mmnga/aixsatoshi-Honyaku-13b-gguf", + "downloads": 390, + "description": "aixsatoshi-Honyaku-13b-ggufaixsatoshiさんが公開しているHonyaku-13bのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 5.086407077388016e-05, - "project_name": "luke-japanese-base" + "score": 5.8352537302981634e-05, + "project_name": "aixsatoshi-Honyaku-13b-gguf" }, { - "url": "https://huggingface.co/inu-ai/dolly-japanese-gpt-1b", - "downloads": 364, - "description": "更新履歴2023年5月7日「oasst1-89k-ja」データセットを追加して対話システムに対応しました。", + "url": "https://huggingface.co/gaianet/Llama-3-8B-Japanese-Instruct-GGUF", + "downloads": 390, + "description": "Llama-3-8B-Japanese-Instruct-GGUFOriginal Modelhaqishen/Llama-3-8B-Japanese-InstructRun with GaianetPrompt template:prompt template: llama-3-chatContext size:chat_ctx_size: 4096Run with GaiaNet:Quick start: https://docs.gaianet.ai/node-guide/quick-startCustomize your node: https://docs.gaianet.ai/node-guide/customizeQuantized GGUF ModelsNameQuant methodBitsSizeUse caseLlama-3-8B-Japanese-Instruct-Q2_K.ggufQ2_K23.18 GBsmallest, significant quality loss - not recommended for most purposesLlama-3-8B-Japanese-I", "source": "Hugging Face", - "score": 5.058612503194639e-05, - "project_name": "dolly-japanese-gpt-1b" + "score": 5.8352537302981634e-05, + "project_name": "Llama-3-8B-Japanese-Instruct-GGUF" }, { - "url": "https://huggingface.co/TheBloke/japanese-stablelm-base-beta-70B-GGUF", - "downloads": 364, - "description": "Chat & support: TheBloke's Discord serverWant to contribute?", + "url": "https://huggingface.co/tohoku-nlp/bert-large-japanese-char-v2", + "downloads": 388, + "description": "BERT large Japanese (character-level tokenization with whole word masking, CC-100 and jawiki-20230102)This is a BERT model pretrained on texts in the Japanese language.", "source": "Hugging Face", - "score": 5.058612503194639e-05, - "project_name": "japanese-stablelm-base-beta-70B-GGUF" + "score": 5.80532935219407e-05, + "project_name": "bert-large-japanese-char-v2" }, { - "url": "https://huggingface.co/mmnga/line-corp-japanese-large-lm-3.6b-instruction-sft-gguf", - "downloads": 364, - "description": "line-corporation/japanese-large-lm-3.6b-instruction-sftline-corporationさんが公開しているjapanese-large-lm-3.6b-instruction-sftのgguf変換版です。", + "url": "https://huggingface.co/mmnga/ELYZA-japanese-CodeLlama-7b-gguf", + "downloads": 386, + "description": "ELYZA-japanese-CodeLlama-7b-ggufELYZAさんが公開しているELYZA-japanese-CodeLlama-7b-instructのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 5.058612503194639e-05, - "project_name": "line-corp-japanese-large-lm-3.6b-instruction-sft-gguf" + "score": 5.775404974089977e-05, + "project_name": "ELYZA-japanese-CodeLlama-7b-gguf" }, { - "url": "https://huggingface.co/retrieva-jp/t5-small-medium", - "downloads": 363, - "description": "Model card for model IDThis is a T5 v1.1 model, pre-trained on a Japanese corpus.", + "url": "https://huggingface.co/llm-book/bert-base-japanese-v3-jsts", + "downloads": 383, + "description": "bert-base-japanese-v3-jsts「大規模言語モデル入門」の第5章で紹介している(意味類似度計算)のモデルです。", "source": "Hugging Face", - "score": 5.04471521609795e-05, - "project_name": "t5-small-medium" + "score": 5.730518406933837e-05, + "project_name": "bert-base-japanese-v3-jsts" }, { - "url": "https://huggingface.co/mmnga/SakanaAI-EvoLLM-JP-v1-7B-gguf", - "downloads": 348, - "description": "SakanaAI-EvoLLM-JP-v1-7B-ggufSakanaAIさんが公開しているEvoLLM-JP-v1-7Bのggufフォーマット変換版です。", + "url": "https://huggingface.co/stabilityai/japanese-stable-diffusion-xl", + "downloads": 382, + "description": "By clicking \"Agree\", you agree to the License Agreement and acknowledge Stability AI's Privacy Policy.", "source": "Hugging Face", - "score": 4.836255909647622e-05, - "project_name": "SakanaAI-EvoLLM-JP-v1-7B-gguf" + "score": 5.7155562178817906e-05, + "project_name": "japanese-stable-diffusion-xl" }, { - "url": "https://huggingface.co/ken11/bert-japanese-ner", - "downloads": 345, - "description": "bert-japanese-nerこのモデルは日本語の固有表現抽出タスクを目的として、京都大学 黒橋・褚・村脇研究室が公開しているBERT日本語Pretrainedモデルをベースにストックマーク株式会社が公開しているner-wikipedia-datasetでファインチューニングしたものです。", + "url": "https://huggingface.co/tsmatz/mt5_summarize_japanese", + "downloads": 379, + "description": "mt5_summarize_japanese(Japanese caption : 日本語の要約のモデル)This model is a fine-tuned version of google/mt5-small trained for Japanese summarization.", "source": "Hugging Face", - "score": 4.794564048357556e-05, - "project_name": "bert-japanese-ner" + "score": 5.6706696507256515e-05, + "project_name": "mt5_summarize_japanese" }, { "url": "https://huggingface.co/alfredplpl/Llama-3-8B-Instruct-Ja", - "downloads": 344, + "downloads": 377, "description": "日本語向け Llama 3 8BはじめにこのリポジトリはLlama 3を日本語化しようとしたモデルのリポジトリです。", "source": "Hugging Face", - "score": 4.780666761260867e-05, + "score": 5.640745272621558e-05, "project_name": "Llama-3-8B-Instruct-Ja" }, { "url": "https://huggingface.co/kit-nlp/bert-base-japanese-sentiment-irony", - "downloads": 341, + "downloads": 354, "description": "BERT Base Japanese for IronyThis is a BERT Base model for sentiment analysis in Japanese additionally finetuned for automatic irony detection.", "source": "Hugging Face", - "score": 4.7389748999708016e-05, + "score": 5.2966149244244866e-05, "project_name": "bert-base-japanese-sentiment-irony" }, { - "url": "https://huggingface.co/ku-nlp/deberta-v2-large-japanese", - "downloads": 339, - "description": "Model Card for Japanese DeBERTa V2 largeModel descriptionThis is a Japanese DeBERTa V2 large model pre-trained on Japanese Wikipedia, the Japanese portion of CC-100, and theJapanese portion of OSCAR.How to useYou can use this model for masked language modeling as follows:from transformers import AutoTokenizer, AutoModelForMaskedLMtokenizer = AutoTokenizer.from_pretrained('ku-nlp/deberta-v2-large-japanese')", + "url": "https://huggingface.co/retrieva-jp/t5-large-long", + "downloads": 349, + "description": "Model card for model IDThis is a T5 v1.1 model, pre-trained on a Japanese corpus.", "source": "Hugging Face", - "score": 4.7111803257774244e-05, - "project_name": "deberta-v2-large-japanese" + "score": 5.221803979164254e-05, + "project_name": "t5-large-long" }, { - "url": "https://huggingface.co/kotoba-tech/kotoba-speech-v0.1", - "downloads": 333, - "description": "Kotoba-Speech-v0.1Kotoba-Speech v0.1 is a 1.2B Transformer-based speech generative model.", + "url": "https://huggingface.co/Fugaku-LLM/Fugaku-LLM-13B-instruct-gguf", + "downloads": 346, + "description": "Fugaku-LLM利用規約この利用規約（以下「本規約」といいます）は、富士通株式会社、国立研究開発法人理化学研究所、国立大学法人東京工業大学、国立大学法人東北大学、株式会社サイバーエージェント、国立大学法人東海国立大学機構、及び株式会社Kotoba Technologies Japan (以下「開発者」といいます)による、スーパーコンピュータ「富岳」政策対応枠における大規模言語モデル分散並列学習手法の開発の成果物として公開する大規模言語モデル（以下「Fugaku-LLM」といいます）の利用に関する条件を定めるものです。", "source": "Hugging Face", - "score": 4.627796603197293e-05, - "project_name": "kotoba-speech-v0.1" + "score": 5.1769174120081145e-05, + "project_name": "Fugaku-LLM-13B-instruct-gguf" }, { - "url": "https://huggingface.co/stabilityai/japanese-stablelm-base-beta-7b", - "downloads": 333, - "description": "Japanese-StableLM-Base-Beta-7BA cute robot wearing a kimono writes calligraphy with one single brush — Stable Diffusion XLModel Descriptionjapanese-stablelm-base-beta-7b is a 7B-parameter decoder-only language model based on Llama-2-7b that has been fine-tuned on a diverse collection of Japanese data, with the intent of maximizing downstream performance on Japanese language tasks.", + "url": "https://huggingface.co/MCZK/ArrowPro-7B-KUJIRA-GGUF", + "downloads": 335, + "description": "DataPilot様の ArrowPro-7B-KUJIRA をGGUF形式に変換したものです。", "source": "Hugging Face", - "score": 4.627796603197293e-05, - "project_name": "japanese-stablelm-base-beta-7b" + "score": 5.012333332435602e-05, + "project_name": "ArrowPro-7B-KUJIRA-GGUF" }, { - "url": "https://huggingface.co/umiyuki/Japanese-WizardLM2-ChatV-7B-GGUF", - "downloads": 330, - "description": "Japanese-WizardLM2-ChatV-7B-GGUFGGUF conversion of \"Japanese-WizardLM2-ChatV-7B\"This model, Japanese-WizardLM2-ChatV-7B, is based on \"chatntq-ja-7b-v1.0 \", and was created by subtracting \"Mistral-7B-v0.1\" from \"WizardLM-2-7b\" ChatVector was added by a factor of 1.0.We aimed to add the high performance of WizardLM-2 to the Japanese language capability of ChatNTQ.", + "url": "https://huggingface.co/studio-ousia/luke-japanese-base", + "downloads": 333, + "description": "luke-japaneseluke-japanese is the Japanese version of LUKE (Language Understanding with Knowledge-based Embeddings), a pre-trained knowledge-enhanced contextualized representation of words and entities.", "source": "Hugging Face", - "score": 4.586104741907228e-05, - "project_name": "Japanese-WizardLM2-ChatV-7B-GGUF" + "score": 4.982408954331509e-05, + "project_name": "luke-japanese-base" }, { - "url": "https://huggingface.co/abeja/Mixtral-8x7B-v0.1-japanese", - "downloads": 326, - "description": "Mixtral-8x7B-v0.1-japaneseMixtral-8x7B-v0.1-japaneseはMixtral-8x7B-v0.1をベースに日本語の語彙拡張継続事前学習を実施したモデルです。", + "url": "https://huggingface.co/MCZK/ArrowPro-7B-RobinHood-GGUF", + "downloads": 331, + "description": "DataPilot様の ArrowPro-7B-RobinHood をGGUF形式に変換したものです。", "source": "Hugging Face", - "score": 4.530515593520473e-05, - "project_name": "Mixtral-8x7B-v0.1-japanese" + "score": 4.9524845762274155e-05, + "project_name": "ArrowPro-7B-RobinHood-GGUF" }, { - "url": "https://huggingface.co/votepurchase/Yaki-Dofu-Mix", - "downloads": 324, - "description": "Yaki-Dofu-Mix概要 / OverviewYaki-Dofu-Mixは、アニメ風の画風に特化したマージモデルです。 ", + "url": "https://huggingface.co/clu-ling/whisper-large-v2-japanese-5k-steps", + "downloads": 327, + "description": "whisper-large-v2-japanese-5k-stepsThis model is a fine-tuned version of openai/whisper-large-v2 on the Japanese CommonVoice dataset (v11)..", "source": "Hugging Face", - "score": 4.502721019327096e-05, - "project_name": "Yaki-Dofu-Mix" + "score": 4.892635820019229e-05, + "project_name": "whisper-large-v2-japanese-5k-steps" }, { - "url": "https://huggingface.co/rinna/nue-asr", - "downloads": 324, - "description": "rinna/nue-asrOverview[Paper][GitHub]We propose a novel end-to-end speech recognition model, Nue ASR, which integrates pre-trained speech and language models.", + "url": "https://huggingface.co/mmnga/Qwen1.5-110B-Chat-gguf", + "downloads": 327, + "description": "Qwen1.5-110B-Chat-ggufQwenさんが公開しているQwen1.5-110B-Chatのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 4.502721019327096e-05, - "project_name": "nue-asr" + "score": 4.892635820019229e-05, + "project_name": "Qwen1.5-110B-Chat-gguf" }, { - "url": "https://huggingface.co/mmnga/shisa-7b-v1-gguf", - "downloads": 312, - "description": "shisa-7b-v1-ggufaugmxntさんが公開しているshisa-7b-v1のggufフォーマット変換版です。", + "url": "https://huggingface.co/classla/xlm-roberta-base-multilingual-text-genre-classifier", + "downloads": 323, + "description": "X-GENRE classifier - multilingual text genre classifierText classification model based on xlm-roberta-base and fine-tuned on a combination of three genre datasets: Slovene GINCO dataset (Kuzman et al.", "source": "Hugging Face", - "score": 4.3359535741668337e-05, - "project_name": "shisa-7b-v1-gguf" + "score": 4.8327870638110434e-05, + "project_name": "xlm-roberta-base-multilingual-text-genre-classifier" }, { - "url": "https://huggingface.co/stockmark/stockmark-13b-instruct", - "downloads": 302, - "description": "Stockmark-13b-instructStockmark-13b-instruct is an instruction-tuned version of Stockmark-13b, a 13 billion parameter Japanese LLM.", - "source": "Hugging Face", - "score": 4.196980703199947e-05, - "project_name": "stockmark-13b-instruct" - }, - { - "url": "https://huggingface.co/dddump/Japanese-TextGen-Kage-v0.1-2x7B-gguf", - "downloads": 302, - "description": "Japanese-TextGen-Kage-v0.1-2x7BKage is \"影\" in Japanese or \"Shadow\" in English.", - "source": "Hugging Face", - "score": 4.196980703199947e-05, - "project_name": "Japanese-TextGen-Kage-v0.1-2x7B-gguf" - }, - { - "url": "https://huggingface.co/mmnga/line-corp-japanese-large-lm-3.6b-gguf", - "downloads": 300, - "description": "line-corporation/japanese-large-lm-3.6bline-corporationさんが公開しているjapanese-large-lm-3.6bのgguf変換版です。", - "source": "Hugging Face", - "score": 4.169186129006571e-05, - "project_name": "line-corp-japanese-large-lm-3.6b-gguf" - }, - { - "url": "https://huggingface.co/MCZK/Japanese-Chat-Umievo-itr004-7b-GGUF", - "downloads": 298, - "description": "umiyuki様の Japanese-Chat-Umievo-itr004-7b をGGUF形式に変換したものです。", - "source": "Hugging Face", - "score": 4.1413915548131935e-05, - "project_name": "Japanese-Chat-Umievo-itr004-7b-GGUF" - }, - { - "url": "https://huggingface.co/OrionStarAI/Orion-14B-Chat-Int4", - "downloads": 292, - "description": "Orion-14B🌐English | 🇨🇳中文 | 🇯🇵日本語 | 🇰🇷한국어🤗", + "url": "https://huggingface.co/mmnga/line-corp-japanese-large-lm-3.6b-instruction-sft-gguf", + "downloads": 321, + "description": "line-corporation/japanese-large-lm-3.6b-instruction-sftline-corporationさんが公開しているjapanese-large-lm-3.6b-instruction-sftのgguf変換版です。", "source": "Hugging Face", - "score": 4.058007832233062e-05, - "project_name": "Orion-14B-Chat-Int4" + "score": 4.80286268570695e-05, + "project_name": "line-corp-japanese-large-lm-3.6b-instruction-sft-gguf" }, { - "url": "https://huggingface.co/MaziyarPanahi/japanese-stablelm-base-gamma-7b-Mistral-7B-Instruct-v0.1-GGUF", - "downloads": 286, - "description": "MaziyarPanahi/japanese-stablelm-base-gamma-7b-Mistral-7B-Instruct-v0.1-GGUFModel creator: MaziyarPanahiOriginal model: MaziyarPanahi/japanese-stablelm-base-gamma-7b-Mistral-7B-Instruct-v0.1DescriptionMaziyarPanahi/japanese-stablelm-base-gamma-7b-Mistral-7B-Instruct-v0.1-GGUF contains GGUF format model files for MaziyarPanahi/japanese-stablelm-base-gamma-7b-Mistral-7B-Instruct-v0.1.How to useThanks to TheBloke for preparing an amazing README on how to use GGUF models:About GGUFGGUF is a new format introduced", + "url": "https://huggingface.co/kotoba-tech/kotoba-speech-v0.1", + "downloads": 314, + "description": "Kotoba-Speech-v0.1Kotoba-Speech v0.1 is a 1.2B Transformer-based speech generative model.", "source": "Hugging Face", - "score": 3.9746241096529306e-05, - "project_name": "japanese-stablelm-base-gamma-7b-Mistral-7B-Instruct-v0.1-GGUF" + "score": 4.698127362342624e-05, + "project_name": "kotoba-speech-v0.1" }, { - "url": "https://huggingface.co/Aratako/Ninja-v1-RP-expressive-breadcrumbs", - "downloads": 280, - "description": "Ninja-v1-RP-expressive-breadcrumbsGGUF版はこちら/Click here for the GGUF version概要This is a merge of pre-trained language models created using mergekit.", + "url": "https://huggingface.co/mmnga/SakanaAI-EvoLLM-JP-A-v1-7B-gguf", + "downloads": 313, + "description": "SakanaAI-EvoLLM-JP-A-v1-7B-ggufSakanaAIさんが公開しているEvoLLM-JP-A-v1-7Bのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 3.8912403870727995e-05, - "project_name": "Ninja-v1-RP-expressive-breadcrumbs" + "score": 4.6831651732905775e-05, + "project_name": "SakanaAI-EvoLLM-JP-A-v1-7B-gguf" }, { - "url": "https://huggingface.co/Tanrei/GPTSAN-japanese", - "downloads": 273, - "description": "Model Card for Tanrei/GPTSAN-japaneseGeneral-purpose Swich transformer based Japanese language modelGPTSAN has some unique features.", + "url": "https://huggingface.co/maddes8cht/stabilityai-japanese-stablelm-3b-4e1t-base-gguf", + "downloads": 311, + "description": "I'm constantly enhancing these model descriptions to provide you with the most relevant and comprehensive informationjapanese-stablelm-3b-4e1t-base - GGUFModel creator: stabilityaiOriginal model: japanese-stablelm-3b-4e1t-baseStableLMThis is a Model based on StableLM.Stablelm is a familiy of Language Models by Stability AI.Note:Current (as of 2023-11-15) implementations of Llama.cpp only support GPU offloading up to 34 Layers with these StableLM Models.", "source": "Hugging Face", - "score": 3.7939593773959794e-05, - "project_name": "GPTSAN-japanese" + "score": 4.653240795186484e-05, + "project_name": "stabilityai-japanese-stablelm-3b-4e1t-base-gguf" }, { - "url": "https://huggingface.co/izumi-lab/deberta-v2-base-japanese", - "downloads": 268, - "description": "DeBERTa V2 base JapaneseThis is a DeBERTaV2 model pretrained on Japanese texts.", + "url": "https://huggingface.co/mmnga/SakanaAI-EvoLLM-JP-v1-7B-gguf", + "downloads": 308, + "description": "SakanaAI-EvoLLM-JP-v1-7B-ggufSakanaAIさんが公開しているEvoLLM-JP-v1-7Bのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 3.7244729419125365e-05, - "project_name": "deberta-v2-base-japanese" + "score": 4.6083542280303445e-05, + "project_name": "SakanaAI-EvoLLM-JP-v1-7B-gguf" }, { - "url": "https://huggingface.co/taoki/phi3-mini-4k-qlora-jmultiwoz-dolly-amenokaku-alpaca_jp_python-GGUF", - "downloads": 266, - "description": "This repository contains a model trained (QLoRA-SFT)", + "url": "https://huggingface.co/maddes8cht/stabilityai-japanese-stablelm-3b-4e1t-instruct-gguf", + "downloads": 307, + "description": "I'm constantly enhancing these model descriptions to provide you with the most relevant and comprehensive informationjapanese-stablelm-3b-4e1t-instruct - GGUFModel creator: stabilityaiOriginal model: japanese-stablelm-3b-4e1t-instructStableLMThis is a Model based on StableLM.Stablelm is a familiy of Language Models by Stability AI.Note:Current (as of 2023-11-15) implementations of Llama.cpp only support GPU offloading up to 34 Layers with these StableLM Models.", "source": "Hugging Face", - "score": 3.696678367719159e-05, - "project_name": "phi3-mini-4k-qlora-jmultiwoz-dolly-amenokaku-alpaca_jp_python-GGUF" + "score": 4.593392038978298e-05, + "project_name": "stabilityai-japanese-stablelm-3b-4e1t-instruct-gguf" }, { - "url": "https://huggingface.co/elyza/ELYZA-japanese-CodeLlama-7b", - "downloads": 260, - "description": "ELYZA-japanese-CodeLlama-7bModel DescriptionELYZA-japanese-CodeLlama-7b は、 Code Llamaをベースとして日本語能力を拡張するために追加事前学習を行ったモデルです。", + "url": "https://huggingface.co/stockmark/stockmark-13b-instruct", + "downloads": 301, + "description": "Stockmark-13b-instructStockmark-13b-instruct is an instruction-tuned version of Stockmark-13b, a 13 billion parameter Japanese LLM.", "source": "Hugging Face", - "score": 3.6132946451390275e-05, - "project_name": "ELYZA-japanese-CodeLlama-7b" + "score": 4.503618904666018e-05, + "project_name": "stockmark-13b-instruct" }, { - "url": "https://huggingface.co/esnya/japanese_speecht5_tts", - "downloads": 260, - "description": "SpeechT5 (TTS task) for JapaneseSpeechT5 model fine-tuned for Japanese speech synthesis (text-to-speech)", + "url": "https://huggingface.co/ku-nlp/gpt2-small-japanese-char", + "downloads": 297, + "description": "Model Card for Japanese character-level GPT-2 SmallModel descriptionThis is a Japanese character-level GPT-2 Small (90M parameters) language model pre-trained on Japanese Wikipedia, the Japanese portion of CC-100, and the Japanese portion of OSCAR.How to useYou can use this model directly with a pipeline for text generation.", "source": "Hugging Face", - "score": 3.6132946451390275e-05, - "project_name": "japanese_speecht5_tts" + "score": 4.443770148457832e-05, + "project_name": "gpt2-small-japanese-char" }, { - "url": "https://huggingface.co/MaziyarPanahi/japanese-stablelm-instruct-gamma-7b-Mistral-7B-Instruct-v0.1-GGUF", - "downloads": 256, - "description": "MaziyarPanahi/japanese-stablelm-instruct-gamma-7b-Mistral-7B-Instruct-v0.1-GGUFModel creator: MaziyarPanahiOriginal model: MaziyarPanahi/japanese-stablelm-instruct-gamma-7b-Mistral-7B-Instruct-v0.1DescriptionMaziyarPanahi/japanese-stablelm-instruct-gamma-7b-Mistral-7B-Instruct-v0.1-GGUF contains GGUF format model files for MaziyarPanahi/japanese-stablelm-instruct-gamma-7b-Mistral-7B-Instruct-v0.1.How to useThanks to TheBloke for preparing an amazing README on how to use GGUF models:About GGUFGGUF is a new f", + "url": "https://huggingface.co/stockmark/gpt-neox-japanese-1.4b", + "downloads": 289, + "description": "stockmark/gpt-neox-japanese-1.4bThis repository provides a GPT-NeoX based model with 1.4B parameters pre-trained on Japanese corpus of about 20B tokens.", "source": "Hugging Face", - "score": 3.5577054967522736e-05, - "project_name": "japanese-stablelm-instruct-gamma-7b-Mistral-7B-Instruct-v0.1-GGUF" + "score": 4.32407263604146e-05, + "project_name": "gpt-neox-japanese-1.4b" }, { - "url": "https://huggingface.co/ku-nlp/gpt2-small-japanese-char", - "downloads": 253, - "description": "Model Card for Japanese character-level GPT-2 SmallModel descriptionThis is a Japanese character-level GPT-2 Small (90M parameters) language model pre-trained on Japanese Wikipedia, the Japanese portion of CC-100, and the Japanese portion of OSCAR.How to useYou can use this model directly with a pipeline for text generation.", + "url": "https://huggingface.co/TheBloke/japanese-stablelm-base-beta-70B-GGUF", + "downloads": 287, + "description": "Chat & support: TheBloke's Discord serverWant to contribute?", "source": "Hugging Face", - "score": 3.516013635462208e-05, - "project_name": "gpt2-small-japanese-char" + "score": 4.2941482579373666e-05, + "project_name": "japanese-stablelm-base-beta-70B-GGUF" }, { "url": "https://huggingface.co/line-corporation/japanese-large-lm-1.7b-instruction-sft", - "downloads": 250, + "downloads": 287, "description": "japanese-large-lm-1.7b-instruction-sftThis repository provides a 1.7B parameters Japanese language model, fine-tuned and trained by LINE Corporation.", "source": "Hugging Face", - "score": 3.474321774172142e-05, + "score": 4.2941482579373666e-05, "project_name": "japanese-large-lm-1.7b-instruction-sft" }, { "url": "https://huggingface.co/Mizuiro-sakura/luke-japanese-base-finetuned-QA", - "downloads": 235, + "downloads": 284, "description": "このモデルはluke-japanese-base-liteをファインチューニングして、Question-Answeringに用いれるようにしたものです。", "source": "Hugging Face", - "score": 3.2658624677218134e-05, + "score": 4.249261690781227e-05, "project_name": "luke-japanese-base-finetuned-QA" }, { - "url": "https://huggingface.co/clu-ling/whisper-large-v2-japanese-5k-steps", - "downloads": 232, - "description": "whisper-large-v2-japanese-5k-stepsThis model is a fine-tuned version of openai/whisper-large-v2 on the Japanese CommonVoice dataset (v11)..", + "url": "https://huggingface.co/haqishen/Llama-3-8B-Japanese-Instruct", + "downloads": 280, + "description": "IntroductionWho am I: Qishen Ha", "source": "Hugging Face", - "score": 3.224170606431748e-05, - "project_name": "whisper-large-v2-japanese-5k-steps" + "score": 4.1894129345730404e-05, + "project_name": "Llama-3-8B-Japanese-Instruct" + }, + { + "url": "https://huggingface.co/taoki/phi3-mini-4k-qlora-jmultiwoz-dolly-amenokaku-alpaca_jp_python-GGUF", + "downloads": 272, + "description": "This repository contains a model trained (QLoRA-SFT)", + "source": "Hugging Face", + "score": 4.0697154221566676e-05, + "project_name": "phi3-mini-4k-qlora-jmultiwoz-dolly-amenokaku-alpaca_jp_python-GGUF" }, { "url": "https://huggingface.co/stabilityai/japanese-stable-vlm", - "downloads": 231, + "downloads": 269, "description": "By clicking \"Agree\", you agree to the License Agreement and acknowledge Stability AI's Privacy Policy.", "source": "Hugging Face", - "score": 3.2102733193350595e-05, + "score": 4.024828855000528e-05, "project_name": "japanese-stable-vlm" }, { - "url": "https://huggingface.co/hotchpotch/japanese-reranker-cross-encoder-small-v1", - "downloads": 224, - "description": "hotchpotch/japanese-reranker-cross-encoder-small-v1日本語で学習させた Reranker (CrossEncoder) シリーズです。", + "url": "https://huggingface.co/sbintuitions/sarashina1-7b", + "downloads": 268, + "description": "Sarashina1-7BThis repository provides Japanese language models trained by SB Intuitions.", "source": "Hugging Face", - "score": 3.1129923096582394e-05, - "project_name": "japanese-reranker-cross-encoder-small-v1" + "score": 4.009866665948481e-05, + "project_name": "sarashina1-7b" }, { - "url": "https://huggingface.co/mmnga/Tanuki-ZeRo-gguf", - "downloads": 223, - "description": "Tanuki-ZeRo-ggufkanhatakeyamaさんが公開しているTanuki-ZeRoのggufフォーマット変換版です。", + "url": "https://huggingface.co/retrieva-jp/t5-small-medium", + "downloads": 259, + "description": "Model card for model IDThis is a T5 v1.1 model, pre-trained on a Japanese corpus.", "source": "Hugging Face", - "score": 3.0990950225615505e-05, - "project_name": "Tanuki-ZeRo-gguf" + "score": 3.8752069644800626e-05, + "project_name": "t5-small-medium" }, { "url": "https://huggingface.co/KoichiYasuoka/deberta-base-japanese-aozora-ud-head", - "downloads": 222, + "downloads": 252, "description": "deberta-base-japanese-aozora-ud-headModel", "source": "Hugging Face", - "score": 3.085197735464862e-05, + "score": 3.7704716411157364e-05, "project_name": "deberta-base-japanese-aozora-ud-head" }, { - "url": "https://huggingface.co/kit-nlp/bert-base-japanese-sentiment-cyberbullying", - "downloads": 221, - "description": "electra-base-cyberbullyingThis is a BERT Base model for the Japanese language finetuned for automatic cyberbullying detection.", + "url": "https://huggingface.co/MCZK/Assistance-7B-GGUF", + "downloads": 252, + "description": "Local-Novel-LLM-project様の Assistance をGGUF形式に変換したものです。", "source": "Hugging Face", - "score": 3.071300448368174e-05, - "project_name": "bert-base-japanese-sentiment-cyberbullying" + "score": 3.7704716411157364e-05, + "project_name": "Assistance-7B-GGUF" }, { - "url": "https://huggingface.co/akiFQC/bert-base-japanese-v3_nli-jsnli", - "downloads": 217, - "description": "Cross-Encoder for Natural Language Inference(NLI) for JapaneseConsidering the results of the JNLI evaluation result, we recommend using akiFQC/bert-base-japanese-v3_nli-jsnli-jnli-jsick for natural language inference in Japanese.", + "url": "https://huggingface.co/mmnga/line-corp-japanese-large-lm-3.6b-gguf", + "downloads": 247, + "description": "line-corporation/japanese-large-lm-3.6bline-corporationさんが公開しているjapanese-large-lm-3.6bのgguf変換版です。", "source": "Hugging Face", - "score": 3.0157112999814193e-05, - "project_name": "bert-base-japanese-v3_nli-jsnli" + "score": 3.6956606958555034e-05, + "project_name": "line-corp-japanese-large-lm-3.6b-gguf" }, { - "url": "https://huggingface.co/rinna/nekomata-14b", - "downloads": 212, - "description": "rinna/nekomata-14bOverviewWe conduct continual pre-training of qwen-14b on 66B tokens from a mixture of Japanese and English datasets.", + "url": "https://huggingface.co/elyza/ELYZA-japanese-CodeLlama-7b", + "downloads": 246, + "description": "ELYZA-japanese-CodeLlama-7bModel DescriptionELYZA-japanese-CodeLlama-7b は、 Code Llamaをベースとして日本語能力を拡張するために追加事前学習を行ったモデルです。", "source": "Hugging Face", - "score": 2.9462248644979765e-05, - "project_name": "nekomata-14b" + "score": 3.680698506803457e-05, + "project_name": "ELYZA-japanese-CodeLlama-7b" }, { - "url": "https://huggingface.co/sbintuitions/sarashina1-7b", - "downloads": 211, - "description": "Sarashina1-7BThis repository provides Japanese language models trained by SB Intuitions.", + "url": "https://huggingface.co/ken11/bert-japanese-ner", + "downloads": 230, + "description": "bert-japanese-nerこのモデルは日本語の固有表現抽出タスクを目的として、京都大学 黒橋・褚・村脇研究室が公開しているBERT日本語Pretrainedモデルをベースにストックマーク株式会社が公開しているner-wikipedia-datasetでファインチューニングしたものです。", "source": "Hugging Face", - "score": 2.932327577401288e-05, - "project_name": "sarashina1-7b" + "score": 3.441303481970712e-05, + "project_name": "bert-japanese-ner" }, { - "url": "https://huggingface.co/nlp-waseda/comet-t5-base-japanese", - "downloads": 211, - "description": "COMET-T5 jaFinetuned T5 on ATOMIC ja using a text-to-text language modeling objective.", + "url": "https://huggingface.co/patrickramos/bert-base-japanese-v2-wrime-fine-tune", + "downloads": 229, + "description": "WRIME-fine-tuned BERT base JapaneseThis model is a Japanese BERTBASE fine-tuned on the WRIME dataset.", "source": "Hugging Face", - "score": 2.932327577401288e-05, - "project_name": "comet-t5-base-japanese" + "score": 3.426341292918665e-05, + "project_name": "bert-base-japanese-v2-wrime-fine-tune" }, { - "url": "https://huggingface.co/toshi456/llava-jp-1.3b-v1.0", - "downloads": 209, - "description": "LLaVA-JP Model CardModel detailModel type:LLaVA-JP is a vision-language model that can converse about input images.", + "url": "https://huggingface.co/hotchpotch/japanese-reranker-cross-encoder-small-v1", + "downloads": 228, + "description": "hotchpotch/japanese-reranker-cross-encoder-small-v1日本語で学習させた Reranker (CrossEncoder) シリーズです。", "source": "Hugging Face", - "score": 2.9045330032079106e-05, - "project_name": "llava-jp-1.3b-v1.0" + "score": 3.411379103866619e-05, + "project_name": "japanese-reranker-cross-encoder-small-v1" }, { - "url": "https://huggingface.co/mmnga/lightblue-Karasu-Mixtral-8x22B-v0.1-gguf", - "downloads": 207, - "description": "lightblue-Karasu-Mixtral-8x22B-v0.1-gguflightblueさんが公開しているKarasu-Mixtral-8x22B-v0.1のggufフォーマット変換版です。", + "url": "https://huggingface.co/turing-motors/heron-chat-git-ja-stablelm-base-7b-v1", + "downloads": 223, + "description": "Heron GIT Japanese StableLM", "source": "Hugging Face", - "score": 2.8767384290145337e-05, - "project_name": "lightblue-Karasu-Mixtral-8x22B-v0.1-gguf" + "score": 3.336568158606386e-05, + "project_name": "heron-chat-git-ja-stablelm-base-7b-v1" }, { "url": "https://huggingface.co/Hemlok/REV-Mix", - "downloads": 206, + "downloads": 222, "description": "◆REV-Mix\"レボリューション\"なモデルです。", "source": "Hugging Face", - "score": 2.862841141917845e-05, + "score": 3.321605969554339e-05, "project_name": "REV-Mix" }, { - "url": "https://huggingface.co/AIBunCho/japanese-novel-gpt-j-6b", - "downloads": 205, - "description": "AIBunCho/japanese-novel-gpt-j-6bAI BunChoで利用しているモデルです。", + "url": "https://huggingface.co/Aratako/Ninja-v1-RP-GGUF", + "downloads": 222, + "description": "Ninja-v1-RP-GGUF概要Aratako/Ninja-v1-RPの量子化済みGGUF版です。", "source": "Hugging Face", - "score": 2.8489438548211564e-05, - "project_name": "japanese-novel-gpt-j-6b" + "score": 3.321605969554339e-05, + "project_name": "Ninja-v1-RP-GGUF" }, { - "url": "https://huggingface.co/stanfordnlp/stanza-ja", - "downloads": 202, - "description": "Stanza model for Japanese (ja)Stanza is a collection of accurate and efficient tools for the linguistic analysis of many human languages.", + "url": "https://huggingface.co/Tanrei/GPTSAN-japanese", + "downloads": 222, + "description": "Model Card for Tanrei/GPTSAN-japaneseGeneral-purpose Swich transformer based Japanese language modelGPTSAN has some unique features.", "source": "Hugging Face", - "score": 2.807251993531091e-05, - "project_name": "stanza-ja" + "score": 3.321605969554339e-05, + "project_name": "GPTSAN-japanese" }, { - "url": "https://huggingface.co/nvidia/parakeet-tdt_ctc-0.6b-ja", - "downloads": 196, - "description": "Parakeet TDT-CTC 0.6B (ja)||parakeet-tdt_ctc-0.6b-ja is an ASR model that transcribes Japanese speech with Punctuations.", + "url": "https://huggingface.co/MCZK/Japanese-Chat-Umievo-itr004-7b-GGUF", + "downloads": 219, + "description": "umiyuki様の Japanese-Chat-Umievo-itr004-7b をGGUF形式に変換したものです。", "source": "Hugging Face", - "score": 2.7238682709509594e-05, - "project_name": "parakeet-tdt_ctc-0.6b-ja" + "score": 3.2767194023981994e-05, + "project_name": "Japanese-Chat-Umievo-itr004-7b-GGUF" }, { - "url": "https://huggingface.co/izumi-lab/bert-base-japanese-fin-additional", - "downloads": 194, - "description": "Additional pretrained BERT base Japanese financeThis is a BERT model pretrained on texts in the Japanese language.", + "url": "https://huggingface.co/umiyuki/Japanese-WizardLM2-ChatV-7B-GGUF", + "downloads": 218, + "description": "Japanese-WizardLM2-ChatV-7B-GGUFGGUF conversion of \"Japanese-WizardLM2-ChatV-7B\"This model, Japanese-WizardLM2-ChatV-7B, is based on \"chatntq-ja-7b-v1.0 \", and was created by subtracting \"Mistral-7B-v0.1\" from \"WizardLM-2-7b\" ChatVector was added by a factor of 1.0.We aimed to add the high performance of WizardLM-2 to the Japanese language capability of ChatNTQ.", "source": "Hugging Face", - "score": 2.696073696757582e-05, - "project_name": "bert-base-japanese-fin-additional" + "score": 3.261757213346153e-05, + "project_name": "Japanese-WizardLM2-ChatV-7B-GGUF" }, { - "url": "https://huggingface.co/umiyuki/Japanese-Chat-Umievo-itr004-7b", - "downloads": 191, - "description": "japanese-chat-umievo-itr004-7bThis is a merge of pre-trained language models created using mergekit.", + "url": "https://huggingface.co/MaziyarPanahi/japanese-stablelm-base-gamma-7b-Mistral-7B-Instruct-v0.1-GGUF", + "downloads": 216, + "description": "MaziyarPanahi/japanese-stablelm-base-gamma-7b-Mistral-7B-Instruct-v0.1-GGUFModel creator: MaziyarPanahiOriginal model: MaziyarPanahi/japanese-stablelm-base-gamma-7b-Mistral-7B-Instruct-v0.1DescriptionMaziyarPanahi/japanese-stablelm-base-gamma-7b-Mistral-7B-Instruct-v0.1-GGUF contains GGUF format model files for MaziyarPanahi/japanese-stablelm-base-gamma-7b-Mistral-7B-Instruct-v0.1.How to useThanks to TheBloke for preparing an amazing README on how to use GGUF models:About GGUFGGUF is a new format introduced", "source": "Hugging Face", - "score": 2.6543818354675166e-05, - "project_name": "Japanese-Chat-Umievo-itr004-7b" + "score": 3.2318328352420596e-05, + "project_name": "japanese-stablelm-base-gamma-7b-Mistral-7B-Instruct-v0.1-GGUF" }, { - "url": "https://huggingface.co/nlp-waseda/roberta-large-japanese", - "downloads": 191, - "description": "nlp-waseda/roberta-large-japaneseModel descriptionThis is a Japanese RoBERTa large model pretrained on Japanese Wikipedia and the Japanese portion of CC-100.How to useYou can use this model for masked language modeling as follows:from transformers import AutoTokenizer, AutoModelForMaskedLMtokenizer = AutoTokenizer.from_pretrained(\"nlp-waseda/roberta-large-japanese\")", + "url": "https://huggingface.co/mmnga/Deepreneur-blue-lizard-gguf", + "downloads": 214, + "description": "Deepreneur-blue-lizard-ggufDeepreneurさんが公開しているblue-lizardのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 2.6543818354675166e-05, - "project_name": "roberta-large-japanese" + "score": 3.2019084571379664e-05, + "project_name": "Deepreneur-blue-lizard-gguf" }, { - "url": "https://huggingface.co/nold/Orion-14B-Base-GGUF", - "downloads": 181, - "description": "Orion-14B🌐English | 🇨🇳中文 | 🇯🇵日本語 |🇰🇷한국어🤗", + "url": "https://huggingface.co/mmnga/shisa-7b-v1-gguf", + "downloads": 214, + "description": "shisa-7b-v1-ggufaugmxntさんが公開しているshisa-7b-v1のggufフォーマット変換版です。", "source": "Hugging Face", - "score": 2.515408964500631e-05, - "project_name": "Orion-14B-Base-GGUF" + "score": 3.2019084571379664e-05, + "project_name": "shisa-7b-v1-gguf" }, { - "url": "https://huggingface.co/izumi-lab/deberta-v2-small-japanese", - "downloads": 180, - "description": "DeBERTa V2 small JapaneseThis is a DeBERTaV2 model pretrained on Japanese texts.", + "url": "https://huggingface.co/nvidia/parakeet-tdt_ctc-0.6b-ja", + "downloads": 213, + "description": "Parakeet TDT-CTC 0.6B (ja)||parakeet-tdt_ctc-0.6b-ja is an ASR model that transcribes Japanese speech with Punctuations.", "source": "Hugging Face", - "score": 2.5015116774039423e-05, - "project_name": "deberta-v2-small-japanese" + "score": 3.18694626808592e-05, + "project_name": "parakeet-tdt_ctc-0.6b-ja" }, { - "url": "https://huggingface.co/kz/mt5base-finetuned-ECC-japanese-small", - "downloads": 180, - "description": "Google's mt5-base fine-tuned in Japanese to solve error detection and correction task.", + "url": "https://huggingface.co/stanfordnlp/stanza-ja", + "downloads": 208, + "description": "Stanza model for Japanese (ja)Stanza is a collection of accurate and efficient tools for the linguistic analysis of many human languages.", "source": "Hugging Face", - "score": 2.5015116774039423e-05, - "project_name": "mt5base-finetuned-ECC-japanese-small" + "score": 3.1121353228256875e-05, + "project_name": "stanza-ja" }, { - "url": "https://huggingface.co/ybelkada/japanese-roberta-question-answering", - "downloads": 178, - "description": "RoBERTa base Japanese - JaQuADDescriptionA Japanese Question Answering model fine-tuned on JaQuAD.Please refer RoBERTa base Japanese for details about the pre-training model.", + "url": "https://huggingface.co/akiFQC/bert-base-japanese-v3_nli-jsnli", + "downloads": 206, + "description": "Cross-Encoder for Natural Language Inference(NLI) for JapaneseConsidering the results of the JNLI evaluation result, we recommend using akiFQC/bert-base-japanese-v3_nli-jsnli-jnli-jsick for natural language inference in Japanese.", "source": "Hugging Face", - "score": 2.473717103210565e-05, - "project_name": "japanese-roberta-question-answering" + "score": 3.082210944721594e-05, + "project_name": "bert-base-japanese-v3_nli-jsnli" }, { - "url": "https://huggingface.co/patrickramos/bert-base-japanese-v2-wrime-fine-tune", - "downloads": 174, - "description": "WRIME-fine-tuned BERT base JapaneseThis model is a Japanese BERTBASE fine-tuned on the WRIME dataset.", + "url": "https://huggingface.co/stabilityai/japanese-stablelm-base-beta-7b", + "downloads": 205, + "description": "Japanese-StableLM-Base-Beta-7BA cute robot wearing a kimono writes calligraphy with one single brush — Stable Diffusion XLModel Descriptionjapanese-stablelm-base-beta-7b is a 7B-parameter decoder-only language model based on Llama-2-7b that has been fine-tuned on a diverse collection of Japanese data, with the intent of maximizing downstream performance on Japanese language tasks.", "source": "Hugging Face", - "score": 2.418127954823811e-05, - "project_name": "bert-base-japanese-v2-wrime-fine-tune" + "score": 3.067248755669548e-05, + "project_name": "japanese-stablelm-base-beta-7b" }, { - "url": "https://huggingface.co/if001/llama2_ja_small", - "downloads": 173, - "description": "日本語でtrainingしたllama2model size: ", + "url": "https://huggingface.co/rinna/nekomata-14b", + "downloads": 203, + "description": "rinna/nekomata-14bOverviewWe conduct continual pre-training of qwen-14b on 66B tokens from a mixture of Japanese and English datasets.", "source": "Hugging Face", - "score": 2.4042306677271222e-05, - "project_name": "llama2_ja_small" + "score": 3.0373243775654545e-05, + "project_name": "nekomata-14b" }, { - "url": "https://huggingface.co/Aratako/Ninja-v1-RP-expressive", - "downloads": 173, - "description": "Ninja-v1-RP-expressiveGGUF版はこちら/Click here for the GGUF version概要This is a merge of pre-trained language models created using mergekit.", + "url": "https://huggingface.co/KoichiYasuoka/bert-base-japanese-wikipedia-ud-head", + "downloads": 203, + "description": "bert-base-japanese-wikipedia-ud-headModel", "source": "Hugging Face", - "score": 2.4042306677271222e-05, - "project_name": "Ninja-v1-RP-expressive" + "score": 3.0373243775654545e-05, + "project_name": "bert-base-japanese-wikipedia-ud-head" }, { - "url": "https://huggingface.co/izumi-lab/bert-small-japanese", - "downloads": 172, - "description": "BERT small Japanese financeThis is a BERT model pretrained on texts in the Japanese language.", + "url": "https://huggingface.co/AIBunCho/japanese-novel-gpt-j-6b", + "downloads": 200, + "description": "AIBunCho/japanese-novel-gpt-j-6bAI BunChoで利用しているモデルです。", "source": "Hugging Face", - "score": 2.3903333806304336e-05, - "project_name": "bert-small-japanese" + "score": 2.9924378104093147e-05, + "project_name": "japanese-novel-gpt-j-6b" }, { - "url": "https://huggingface.co/Kendamarron/fineweb-edu-classifier-ja-v2", - "downloads": 172, - "description": "HuggingFaceFW/fineweb-edu-classifierを再現するために、日本語データでtohoku-nlp/bert-base-japanese-v3を学習したモデルです。", + "url": "https://huggingface.co/MCZK/Tora-7B-v0.2-GGUF", + "downloads": 196, + "description": "ryota39様の Tora-7B-v0.2 をGGUF形式に変換したものです。", "source": "Hugging Face", - "score": 2.3903333806304336e-05, - "project_name": "fineweb-edu-classifier-ja-v2" + "score": 2.9325890542011283e-05, + "project_name": "Tora-7B-v0.2-GGUF" }, { - "url": "https://huggingface.co/cinmodel/electra-small-japanese-generator", - "downloads": 168, - "description": "Japanese ELECTRA-smallWe provide a Japanese ELECTRA-Small model, as described in ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators.", + "url": "https://huggingface.co/izumi-lab/bert-base-japanese-fin-additional", + "downloads": 195, + "description": "Additional pretrained BERT base Japanese financeThis is a BERT model pretrained on texts in the Japanese language.", "source": "Hugging Face", - "score": 2.3347442322436794e-05, - "project_name": "electra-small-japanese-generator" + "score": 2.9176268651490817e-05, + "project_name": "bert-base-japanese-fin-additional" + }, + { + "url": "https://huggingface.co/nlp-waseda/comet-gpt2-small-japanese", + "downloads": 192, + "description": "COMET-GPT2 jaFinetuned GPT-2 on ATOMIC ja using a causal language modeling (CLM) objective.", + "source": "Hugging Face", + "score": 2.872740297992942e-05, + "project_name": "comet-gpt2-small-japanese" }, { "url": "https://huggingface.co/retrieva-jp/t5-large-short", - "downloads": 167, + "downloads": 192, "description": "Model card for model IDThis is a T5 v1.1 model, pre-trained on a Japanese corpus.", "source": "Hugging Face", - "score": 2.3208469451469908e-05, + "score": 2.872740297992942e-05, "project_name": "t5-large-short" }, { - "url": "https://huggingface.co/natsusakiyomi/KaedeMix", - "downloads": 158, - "description": "📄 ライセンス / License修正 CreativeML OpenRAIL-M ライセンス / Modified CreativeML OpenRAIL-M licenseこのモデルのクレジットを入れずに使用するUse the model without crediting the creatorこのモデルで生成した画像を商用利用するSell images they generateこのモデルを商用の画像生成サービスで利用するRun on services that generate images for moneyこのモデルを使用したマージモデルを共有するShare merges using this modelこのモデル、またはこのモデルをマージしたモデルを販売するSell this model or merges using this modelこのモデルをマージしたモデルに異なる権限を設定するHave different permissions when sharing merges🖼️ 例 / Examples(※他の人が生成した物を表示している場合は本人の許諾を得て表示しています)���ちPさん作", + "url": "https://huggingface.co/nlp-waseda/comet-t5-base-japanese", + "downloads": 186, + "description": "COMET-T5 jaFinetuned T5 on ATOMIC ja using a text-to-text language modeling objective.", "source": "Hugging Face", - "score": 2.1957713612767938e-05, - "project_name": "KaedeMix" + "score": 2.7829671636806627e-05, + "project_name": "comet-t5-base-japanese" }, { - "url": "https://huggingface.co/OrionStarAI/Orion-14B-LongChat", - "downloads": 156, - "description": "Orion-14B🌐English | 🇨🇳中文 | 🇯🇵日本語 | 🇰🇷한국어🤗", + "url": "https://huggingface.co/Aratako/c4ai-command-r-v01-japanese-instruct-GGUF", + "downloads": 186, + "description": "c4ai-command-r-v01-japanese-instruct-GGUF概要Aratako/c4ai-command-r-v01-japanese-instructの量子化済みGGUF版です。", "source": "Hugging Face", - "score": 2.1679767870834168e-05, - "project_name": "Orion-14B-LongChat" + "score": 2.7829671636806627e-05, + "project_name": "c4ai-command-r-v01-japanese-instruct-GGUF" }, { - "url": "https://huggingface.co/oshizo/qa-refine-japanese-gpt-1b", - "downloads": 155, - "description": "Model Card for Model IDこのモデルはrinna/japanese-gpt-1bをベースモデルとして、コンテキストからの抽出型QAと、解答を新たなコンテキストでリファインするための学習を行ったモデルです。", + "url": "https://huggingface.co/MaziyarPanahi/japanese-stablelm-instruct-gamma-7b-Mistral-7B-Instruct-v0.1-GGUF", + "downloads": 186, + "description": "MaziyarPanahi/japanese-stablelm-instruct-gamma-7b-Mistral-7B-Instruct-v0.1-GGUFModel creator: MaziyarPanahiOriginal model: MaziyarPanahi/japanese-stablelm-instruct-gamma-7b-Mistral-7B-Instruct-v0.1DescriptionMaziyarPanahi/japanese-stablelm-instruct-gamma-7b-Mistral-7B-Instruct-v0.1-GGUF contains GGUF format model files for MaziyarPanahi/japanese-stablelm-instruct-gamma-7b-Mistral-7B-Instruct-v0.1.How to useThanks to TheBloke for preparing an amazing README on how to use GGUF models:About GGUFGGUF is a new f", "source": "Hugging Face", - "score": 2.1540794999867282e-05, - "project_name": "qa-refine-japanese-gpt-1b" + "score": 2.7829671636806627e-05, + "project_name": "japanese-stablelm-instruct-gamma-7b-Mistral-7B-Instruct-v0.1-GGUF" }, { - "url": "https://huggingface.co/pfnet/plamo-13b-instruct", - "downloads": 150, - "description": "PLaMo-13B-InstructModel DescriptionPLaMo-13B-Instruct is an instruct fine-tuned model built upon the 8192 context length version of PLaMo-13B text generation model.", + "url": "https://huggingface.co/MCZK/Tora-7B-v0.1-GGUF", + "downloads": 185, + "description": "ryota39様の Tora-7B-v0.1 をGGUF形式に変換したものです。", "source": "Hugging Face", - "score": 2.0845930645032854e-05, - "project_name": "plamo-13b-instruct" + "score": 2.768004974628616e-05, + "project_name": "Tora-7B-v0.1-GGUF" }, { - "url": "https://huggingface.co/MCZK/Tora-7B-v0.2-GGUF", - "downloads": 139, - "description": "ryota39様の Tora-7B-v0.2 をGGUF形式に変換したものです。", + "url": "https://huggingface.co/toshi456/llava-jp-1.3b-v1.0", + "downloads": 181, + "description": "LLaVA-JP Model CardModel detailModel type:LLaVA-JP is a vision-language model that can converse about input images.", "source": "Hugging Face", - "score": 1.931722906439711e-05, - "project_name": "Tora-7B-v0.2-GGUF" + "score": 2.7081562184204297e-05, + "project_name": "llava-jp-1.3b-v1.0" }, { - "url": "https://huggingface.co/reazon-research/reazonspeech-espnet-next", - "downloads": 139, - "description": "reazonspeech-espnet-nextReazonSpeech is a project to maintain freely-available Japanese audiodatasets and ML models.reazonspeech-espnet-next is a \"bleeding-edge\" repository that containslatest ASR models trained by ReazonSpeech team.", + "url": "https://huggingface.co/kz/mt5base-finetuned-ECC-japanese-small", + "downloads": 179, + "description": "Google's mt5-base fine-tuned in Japanese to solve error detection and correction task.", "source": "Hugging Face", - "score": 1.931722906439711e-05, - "project_name": "reazonspeech-espnet-next" + "score": 2.6782318403163365e-05, + "project_name": "mt5base-finetuned-ECC-japanese-small" + }, + { + "url": "https://huggingface.co/izumi-lab/deberta-v2-small-japanese", + "downloads": 177, + "description": "DeBERTa V2 small JapaneseThis is a DeBERTaV2 model pretrained on Japanese texts.", + "source": "Hugging Face", + "score": 2.6483074622122433e-05, + "project_name": "deberta-v2-small-japanese" }, { "url": "https://huggingface.co/tsmatz/roberta_qa_japanese", - "downloads": 139, + "downloads": 177, "description": "roberta_qa_japanese(Japanese caption : 日本語の (抽出型) 質問応答のモデル)This model is a fine-tuned version of rinna/japanese-roberta-base (pre-trained RoBERTa model provided by rinna Co.", "source": "Hugging Face", - "score": 1.931722906439711e-05, + "score": 2.6483074622122433e-05, "project_name": "roberta_qa_japanese" }, + { + "url": "https://huggingface.co/if001/llama2_ja_small", + "downloads": 174, + "description": "日本語でtrainingしたllama2model size: ", + "source": "Hugging Face", + "score": 2.603420895056104e-05, + "project_name": "llama2_ja_small" + }, + { + "url": "https://huggingface.co/Kendamarron/fineweb-edu-classifier-ja-v2", + "downloads": 173, + "description": "HuggingFaceFW/fineweb-edu-classifierを再現するために、日本語データでtohoku-nlp/bert-base-japanese-v3を学習したモデルです。", + "source": "Hugging Face", + "score": 2.5884587060040572e-05, + "project_name": "fineweb-edu-classifier-ja-v2" + }, + { + "url": "https://huggingface.co/cinmodel/electra-small-japanese-generator", + "downloads": 171, + "description": "Japanese ELECTRA-smallWe provide a Japanese ELECTRA-Small model, as described in ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators.", + "source": "Hugging Face", + "score": 2.558534327899964e-05, + "project_name": "electra-small-japanese-generator" + }, + { + "url": "https://huggingface.co/dddump/Japanese-TextGen-Kage-v0.1-2x7B-gguf", + "downloads": 168, + "description": "Japanese-TextGen-Kage-v0.1-2x7BKage is \"影\" in Japanese or \"Shadow\" in English.", + "source": "Hugging Face", + "score": 2.5136477607438243e-05, + "project_name": "Japanese-TextGen-Kage-v0.1-2x7B-gguf" + }, { "url": "https://huggingface.co/nitky/Oumuamua-7b-instruct-v2", - "downloads": 136, + "downloads": 167, "description": "Oumuamua-7b-instruct-v2🚨 If you want to avoid outputs that appear to be literal translations, please prompt this model to role-play as a Japanese person.", "source": "Hugging Face", - "score": 1.8900310451496452e-05, + "score": 2.4986855716917777e-05, "project_name": "Oumuamua-7b-instruct-v2" }, { - "url": "https://huggingface.co/TFMC/ChatNTQ-JA-7b-v1.0-GGUF", - "downloads": 135, - "description": "GGUF conversion of NTQAI/chatntq-ja-7b-v1.0ChatNTQ-JA-7b-v1.0 is a Japanese chat fine-tuned model built on top of the stabilityai/japanese-stablelm-base-gamma-7b, which is originally based on Mistral 7B v0.1.", + "url": "https://huggingface.co/mmnga/Tanuki-ZeRo-gguf", + "downloads": 163, + "description": "Tanuki-ZeRo-ggufkanhatakeyamaさんが公開しているTanuki-ZeRoのggufフォーマット変換版です。", "source": "Hugging Face", - "score": 1.8761337580529566e-05, - "project_name": "ChatNTQ-JA-7b-v1.0-GGUF" + "score": 2.4388368154835913e-05, + "project_name": "Tanuki-ZeRo-gguf" }, { - "url": "https://huggingface.co/turing-motors/heron-chat-blip-ja-stablelm-base-7b-v1-llava-620k", - "downloads": 134, - "description": "Heron BLIP Japanese StableLM", + "url": "https://huggingface.co/oshizo/qa-refine-japanese-gpt-1b", + "downloads": 162, + "description": "Model Card for Model IDこのモデルはrinna/japanese-gpt-1bをベースモデルとして、コンテキストからの抽出型QAと、解答を新たなコンテキストでリファインするための学習を行ったモデルです。", "source": "Hugging Face", - "score": 1.8622364709562683e-05, - "project_name": "heron-chat-blip-ja-stablelm-base-7b-v1-llava-620k" + "score": 2.423874626431545e-05, + "project_name": "qa-refine-japanese-gpt-1b" }, { - "url": "https://huggingface.co/sonoisa/t5-base-japanese-v1.1", - "downloads": 133, - "description": "日本語T5事前学習済みモデルThis is a T5 (Text-to-Text Transfer Transformer) model pretrained on Japanese corpus.", + "url": "https://huggingface.co/ybelkada/japanese-roberta-question-answering", + "downloads": 157, + "description": "RoBERTa base Japanese - JaQuADDescriptionA Japanese Question Answering model fine-tuned on JaQuAD.Please refer RoBERTa base Japanese for details about the pre-training model.", "source": "Hugging Face", - "score": 1.8483391838595796e-05, - "project_name": "t5-base-japanese-v1.1" + "score": 2.349063681171312e-05, + "project_name": "japanese-roberta-question-answering" + }, + { + "url": "https://huggingface.co/mmnga/lightblue-Karasu-Mixtral-8x22B-v0.1-gguf", + "downloads": 155, + "description": "lightblue-Karasu-Mixtral-8x22B-v0.1-gguflightblueさんが公開しているKarasu-Mixtral-8x22B-v0.1のggufフォーマット変換版です。", + "source": "Hugging Face", + "score": 2.319139303067219e-05, + "project_name": "lightblue-Karasu-Mixtral-8x22B-v0.1-gguf" + }, + { + "url": "https://huggingface.co/izumi-lab/bert-small-japanese", + "downloads": 153, + "description": "BERT small Japanese financeThis is a BERT model pretrained on texts in the Japanese language.", + "source": "Hugging Face", + "score": 2.2892149249631256e-05, + "project_name": "bert-small-japanese" + }, + { + "url": "https://huggingface.co/Aratako/Ninja-v1-RP-expressive", + "downloads": 150, + "description": "Ninja-v1-RP-expressiveGGUF版はこちら/Click here for the GGUF version概要This is a merge of pre-trained language models created using mergekit.", + "source": "Hugging Face", + "score": 2.244328357806986e-05, + "project_name": "Ninja-v1-RP-expressive" + }, + { + "url": "https://huggingface.co/sonoisa/sentence-bert-base-ja-en-mean-tokens", + "downloads": 149, + "description": "This is a Japanese+English sentence-BERT model.", + "source": "Hugging Face", + "score": 2.2293661687549392e-05, + "project_name": "sentence-bert-base-ja-en-mean-tokens" + }, + { + "url": "https://huggingface.co/kit-nlp/bert-base-japanese-sentiment-cyberbullying", + "downloads": 147, + "description": "electra-base-cyberbullyingThis is a BERT Base model for the Japanese language finetuned for automatic cyberbullying detection.", + "source": "Hugging Face", + "score": 2.1994417906508464e-05, + "project_name": "bert-base-japanese-sentiment-cyberbullying" + }, + { + "url": "https://huggingface.co/nold/Orion-14B-Base-GGUF", + "downloads": 143, + "description": "Orion-14B🌐English | 🇨🇳中文 | 🇯🇵日本語 |🇰🇷한국어🤗", + "source": "Hugging Face", + "score": 2.13959303444266e-05, + "project_name": "Orion-14B-Base-GGUF" }, { "url": "https://huggingface.co/retrieva-jp/t5-base-medium", - "downloads": 132, + "downloads": 141, "description": "Model card for model IDThis is a T5 v1.1 model, pre-trained on a Japanese corpus.", "source": "Hugging Face", - "score": 1.834441896762891e-05, + "score": 2.1096686563385668e-05, "project_name": "t5-base-medium" }, { - "url": "https://huggingface.co/Local-Novel-LLM-project/Ninja-v1", - "downloads": 131, - "description": "Our ModelsVecteusNinja-v1Ninja-v1-NSFWNinja-v1-128kNinja-v1-NSFW-128kModel Card for Ninja-v1.0The Mistral-7B--based Large Language Model (LLM) is an noveldataset fine-tuned version of the Mistral-7B-v0.1Ninja has the following changes compared to Mistral-7B-v0.1.Achieving both high quality Japanese and English generationMemory ability that does not forget even after long-context generationThis model was created with the help of GPUs from the first LocalAI hackathon.", + "url": "https://huggingface.co/reazon-research/reazonspeech-espnet-next", + "downloads": 138, + "description": "reazonspeech-espnet-nextReazonSpeech is a project to maintain freely-available Japanese audiodatasets and ML models.reazonspeech-espnet-next is a \"bleeding-edge\" repository that containslatest ASR models trained by ReazonSpeech team.", "source": "Hugging Face", - "score": 1.8205446096662024e-05, - "project_name": "Ninja-v1" + "score": 2.064782089182427e-05, + "project_name": "reazonspeech-espnet-next" }, { - "url": "https://huggingface.co/turing-motors/heron-chat-git-ja-stablelm-base-7b-v1", - "downloads": 130, - "description": "Heron GIT Japanese StableLM", + "url": "https://huggingface.co/pfnet/plamo-13b-instruct", + "downloads": 138, + "description": "PLaMo-13B-InstructModel DescriptionPLaMo-13B-Instruct is an instruct fine-tuned model built upon the 8192 context length version of PLaMo-13B text generation model.", "source": "Hugging Face", - "score": 1.8066473225695137e-05, - "project_name": "heron-chat-git-ja-stablelm-base-7b-v1" + "score": 2.064782089182427e-05, + "project_name": "plamo-13b-instruct" }, { - "url": "https://huggingface.co/rinna/nekomata-14b-instruction-gguf", + "url": "https://huggingface.co/natsusakiyomi/AnzuMix", + "downloads": 136, + "description": "AnzuMixSeriesVAEの内臓はないぞ！と言わせないぞ！！！！", + "source": "Hugging Face", + "score": 2.0348577110783338e-05, + "project_name": "AnzuMix" + }, + { + "url": "https://huggingface.co/TheBloke/japanese-stablelm-instruct-beta-70B-GPTQ", "downloads": 130, - "description": "rinna/nekomata-14b-instruction-ggufOverviewThe model is the GGUF version of rinna/nekomata-14b-instruction.", + "description": "Chat & support: TheBloke's Discord serverWant to contribute?", "source": "Hugging Face", - "score": 1.8066473225695137e-05, - "project_name": "nekomata-14b-instruction-gguf" + "score": 1.9450845767660546e-05, + "project_name": "japanese-stablelm-instruct-beta-70B-GPTQ" }, { - "url": "https://huggingface.co/sambanovasystems/SambaLingo-Japanese-Chat", + "url": "https://huggingface.co/turing-motors/heron-chat-blip-ja-stablelm-base-7b-v1-llava-620k", "downloads": 128, - "description": "SambaLingo-Japanese-ChatSambaLingo-Japanese-Chat is a human aligned chat model trained in Japanese and English.", + "description": "Heron BLIP Japanese StableLM", "source": "Hugging Face", - "score": 1.7788527483761368e-05, - "project_name": "SambaLingo-Japanese-Chat" + "score": 1.9151601986619614e-05, + "project_name": "heron-chat-blip-ja-stablelm-base-7b-v1-llava-620k" }, { - "url": "https://huggingface.co/MCZK/Tora-7B-v0.1-GGUF", + "url": "https://huggingface.co/sonoisa/t5-base-japanese-v1.1", "downloads": 126, - "description": "ryota39様の Tora-7B-v0.1 をGGUF形式に変換したものです。", + "description": "日本語T5事前学習済みモデルThis is a T5 (Text-to-Text Transfer Transformer) model pretrained on Japanese corpus.", "source": "Hugging Face", - "score": 1.7510581741827595e-05, - "project_name": "Tora-7B-v0.1-GGUF" + "score": 1.8852358205578682e-05, + "project_name": "t5-base-japanese-v1.1" }, { - "url": "https://huggingface.co/watashiha/watashiha-gpt-6b", + "url": "https://huggingface.co/sonoisa/t5-base-english-japanese", "downloads": 126, - "description": "モデル概要AWSのtrn1インスタンスを用いて開発した大喜利言語モデルです。", + "description": "英語+日本語T5事前学習済みモデルThis is a T5 (Text-to-Text Transfer Transformer) model pretrained on English and Japanese balanced corpus.", "source": "Hugging Face", - "score": 1.7510581741827595e-05, - "project_name": "watashiha-gpt-6b" + "score": 1.8852358205578682e-05, + "project_name": "t5-base-english-japanese" }, { - "url": "https://huggingface.co/abhishek/autonlp-japanese-sentiment-59363", - "downloads": 124, - "description": "Model Trained Using AutoNLPProblem type: Binary ClassificationModel ID: 59363Validation MetricsLoss: 0.12651239335536957Accuracy: 0.9532079853817648Precision: 0.9729688278823665Recall: 0.9744633462616643AUC: 0.9717333684823413F1: 0.9737155136027014UsageYou can use cURL to access this model:$ curl -X POST -H \"Authorization: Bearer YOUR_API_KEY\" -H \"Content-Type: application/json\" -d '{\"inputs\": \"I love AutoNLP\"}'", + "url": "https://huggingface.co/colorfulscoop/gpt2-small-ja", + "downloads": 125, + "description": "GPT-2 small Japanese modelThis repository contains a GPT2-small model trained on Japanese Wikipedia dataset.", "source": "Hugging Face", - "score": 1.7232635999893826e-05, - "project_name": "autonlp-japanese-sentiment-59363" + "score": 1.8702736315058216e-05, + "project_name": "gpt2-small-ja" }, { - "url": "https://huggingface.co/LoneStriker/SambaLingo-Japanese-Chat-GGUF", - "downloads": 121, + "url": "https://huggingface.co/sambanovasystems/SambaLingo-Japanese-Chat", + "downloads": 125, "description": "SambaLingo-Japanese-ChatSambaLingo-Japanese-Chat is a human aligned chat model trained in Japanese and English.", "source": "Hugging Face", - "score": 1.6815717386993167e-05, - "project_name": "SambaLingo-Japanese-Chat-GGUF" + "score": 1.8702736315058216e-05, + "project_name": "SambaLingo-Japanese-Chat" }, { - "url": "https://huggingface.co/colorfulscoop/gpt2-small-ja", - "downloads": 117, - "description": "GPT-2 small Japanese modelThis repository contains a GPT2-small model trained on Japanese Wikipedia dataset.", + "url": "https://huggingface.co/tohoku-nlp/stable-diffusion-xl-jp-base-1.0", + "downloads": 122, + "description": "(English part follows Japanese one.", "source": "Hugging Face", - "score": 1.6259825903125625e-05, - "project_name": "gpt2-small-ja" + "score": 1.8253870643496818e-05, + "project_name": "stable-diffusion-xl-jp-base-1.0" }, { - "url": "https://huggingface.co/ThePioneer/MoeDiffusionPlusPlus", - "downloads": 116, - "description": "モデル説明 (model explanation)V1 = MoeDiffusion 1.0 + (HassanBlend 1.5 - VMix03) * 0.2V2 = MoeDiffusion 0.6 : HassanBlend 1.5 0.2 : VMix03 : 0.2マージ元のルーツにNAIリークやInsta系モデルが含まれるという噂があるので、NAIリークアンチ・Insta系モデル���ンチには非推奨理想の黒髪ポニテ顔が出せるYaguruMagikuを、ある程度顔が近くて制御しやすいAbyssOrangeMix2と混ぜてみた。", + "url": "https://huggingface.co/HODACHI/glm-4-9b-chat-FT-ja-v0.3", + "downloads": 118, + "description": "概要GLM-4-9B-Chatを、日本語のWikiデータを選定し、追加学習した日本語に非常に強いスコアを出したモデルです。", "source": "Hugging Face", - "score": 1.612085303215874e-05, - "project_name": "MoeDiffusionPlusPlus" + "score": 1.7655383081414958e-05, + "project_name": "glm-4-9b-chat-FT-ja-v0.3" }, { - "url": "https://huggingface.co/retrieva-jp/t5-xl", - "downloads": 116, - "description": "Model card for model IDThis is a T5 v1.1 model, pre-trained on a Japanese corpus.", + "url": "https://huggingface.co/rinna/nekomata-14b-instruction-gguf", + "downloads": 117, + "description": "rinna/nekomata-14b-instruction-ggufOverviewThe model is the GGUF version of rinna/nekomata-14b-instruction.", "source": "Hugging Face", - "score": 1.612085303215874e-05, - "project_name": "t5-xl" + "score": 1.750576119089449e-05, + "project_name": "nekomata-14b-instruction-gguf" }, { - "url": "https://huggingface.co/stabilityai/japanese-stablelm-instruct-ja_vocab-beta-7b", - "downloads": 115, - "description": "Japanese-StableLM-Instruct-JAVocab-Beta-7BA cute robot wearing a kimono writes calligraphy with one single brush — Stable Diffusion XLModel Descriptionjapanese-stablelm-instruct-ja_vocab-beta-7b is a 7B-parameter decoder-only language model based on japanese-stablelm-ja_vocab-beta-7b and further fine tuned on Databricks Dolly-15k, Anthropic HH, and other public data.", + "url": "https://huggingface.co/pfnet/plamo-13b-instruct-nc", + "downloads": 116, + "description": "PLaMo-13B-Instruct-NCModel DescriptionPLaMo-13B-Instruct-NC is a noncommercial instruct fine-tuned model built upon the 8192 context length version of PLaMo-13B text generation model.", "source": "Hugging Face", - "score": 1.5981880161191853e-05, - "project_name": "japanese-stablelm-instruct-ja_vocab-beta-7b" + "score": 1.7356139300374026e-05, + "project_name": "plamo-13b-instruct-nc" }, { "url": "https://huggingface.co/ku-nlp/roberta-base-japanese-char-wwm", - "downloads": 114, + "downloads": 115, "description": "ku-nlp/roberta-base-japanese-char-wwmModel descriptionThis is a Japanese RoBERTa base model pre-trained on Japanese Wikipedia and the Japanese portion of CC-100.This model is trained with character-level tokenization and whole word masking.", "source": "Hugging Face", - "score": 1.5842907290224966e-05, + "score": 1.720651740985356e-05, "project_name": "roberta-base-japanese-char-wwm" }, { - "url": "https://huggingface.co/sonoisa/t5-base-english-japanese", - "downloads": 113, - "description": "英語+日本語T5事前学習済みモデルThis is a T5 (Text-to-Text Transfer Transformer) model pretrained on English and Japanese balanced corpus.", + "url": "https://huggingface.co/watashiha/watashiha-gpt-6b", + "downloads": 111, + "description": "モデル概要AWSのtrn1インスタンスを用いて開発した大喜利言語モデルです。", "source": "Hugging Face", - "score": 1.5703934419258083e-05, - "project_name": "t5-base-english-japanese" + "score": 1.6608029847771696e-05, + "project_name": "watashiha-gpt-6b" + }, + { + "url": "https://huggingface.co/sonoisa/sentence-t5-base-ja-mean-tokens", + "downloads": 111, + "description": "This is a Japanese sentence-T5 model.", + "source": "Hugging Face", + "score": 1.6608029847771696e-05, + "project_name": "sentence-t5-base-ja-mean-tokens" + }, + { + "url": "https://huggingface.co/retrieva-jp/t5-xl", + "downloads": 111, + "description": "Model card for model IDThis is a T5 v1.1 model, pre-trained on a Japanese corpus.", + "source": "Hugging Face", + "score": 1.6608029847771696e-05, + "project_name": "t5-xl" }, { "url": "https://huggingface.co/recruit-jp/japanese-typo-detector-roberta-base", - "downloads": 112, + "downloads": 108, "description": "recruit-jp/japanese-typo-detector-roberta-baseモデルの概要日本語の文章を入力すると各文字ごとに誤字脱字である確率を出力します各ラベルの意味は以下の通りですidlabelmeaning0OK誤字なし1deletion1文字の抜け2insertion_a余分な1文字の挿入3insertion_b直前の文字列と一致する２文字以上の余分な文字の挿入4kanji-conversion_a同一の読みを持つ漢字の入れ替え（誤変換）5kanji-conversion_b近い読みを持つ漢字の入れ替え（誤変換）6substitution1文字の入れ替え7transposition隣接する２文字間の転置8othersその他の入力誤り誤り種類の詳細については学習データセットの元論文をご参照ください日本語 Wikipedia の編集履歴に基づく 入力誤りデータセットと訂正システムの改良その他、モデルの詳細については当社ブログ記事をご参照ください誤字脱字検出モデルをHugging Face Hubに公開しました (Recruit Data Blog)学習データ京都大学大学院情報学研究科知能情", "source": "Hugging Face", - "score": 1.5564961548291197e-05, + "score": 1.6159164176210298e-05, "project_name": "japanese-typo-detector-roberta-base" }, { - "url": "https://huggingface.co/HODACHI/glm-4-9b-chat-FT-ja-v0.3", - "downloads": 111, - "description": "概要GLM-4-9B-Chatを、日本語のWikiデータを選定し、追加学習した日本語に非常に強いスコアを出したモデルです。", + "url": "https://huggingface.co/abhishek/autonlp-japanese-sentiment-59363", + "downloads": 108, + "description": "Model Trained Using AutoNLPProblem type: Binary ClassificationModel ID: 59363Validation MetricsLoss: 0.12651239335536957Accuracy: 0.9532079853817648Precision: 0.9729688278823665Recall: 0.9744633462616643AUC: 0.9717333684823413F1: 0.9737155136027014UsageYou can use cURL to access this model:$ curl -X POST -H \"Authorization: Bearer YOUR_API_KEY\" -H \"Content-Type: application/json\" -d '{\"inputs\": \"I love AutoNLP\"}'", "source": "Hugging Face", - "score": 1.542598867732431e-05, - "project_name": "glm-4-9b-chat-FT-ja-v0.3" + "score": 1.6159164176210298e-05, + "project_name": "autonlp-japanese-sentiment-59363" }, { - "url": "https://huggingface.co/stabilityai/japanese-instructblip-alpha", - "downloads": 110, - "description": "Japanese InstructBLIP AlphaModel DetailsJapanese InstructBLIP Alpha is a vision-language instruction-following model that enables to generate Japanese descriptions for input images and optionally input texts such as questions.", + "url": "https://huggingface.co/stabilityai/japanese-stablelm-instruct-ja_vocab-beta-7b", + "downloads": 107, + "description": "Japanese-StableLM-Instruct-JAVocab-Beta-7BA cute robot wearing a kimono writes calligraphy with one single brush — Stable Diffusion XLModel Descriptionjapanese-stablelm-instruct-ja_vocab-beta-7b is a 7B-parameter decoder-only language model based on japanese-stablelm-ja_vocab-beta-7b and further fine tuned on Databricks Dolly-15k, Anthropic HH, and other public data.", "source": "Hugging Face", - "score": 1.5287015806357424e-05, - "project_name": "japanese-instructblip-alpha" + "score": 1.6009542285689832e-05, + "project_name": "japanese-stablelm-instruct-ja_vocab-beta-7b" }, { - "url": "https://huggingface.co/nlp-waseda/bigbird-base-japanese", - "downloads": 104, - "description": "nlp-waseda/bigbird-base-japaneseModel descriptionThis is a Japanese BigBird base model pretrained on Japanese Wikipedia, the Japanese portion of CC-100, and the Japanese portion of OSCAR.How to useYou can use this model for masked language modeling as follows:from transformers import AutoTokenizer, AutoModelForMaskedLMtokenizer = AutoTokenizer.from_pretrained(\"nlp-waseda/bigbird-base-japanese\")", + "url": "https://huggingface.co/oshizo/japanese-e5-mistral-1.9b", + "downloads": 107, + "description": "Model trained on 800,000 Japanese sentences after reducing oshizo/japanese-e5-mistral-7b_slerp to 8 layers.", "source": "Hugging Face", - "score": 1.4453178580556112e-05, - "project_name": "bigbird-base-japanese" + "score": 1.6009542285689832e-05, + "project_name": "japanese-e5-mistral-1.9b" }, { - "url": "https://huggingface.co/sazyou-roukaku/AfterRealXL", - "downloads": 101, - "description": "こちらでアップロードできないので、civitaiにて先に公開しています。", + "url": "https://huggingface.co/natsusakiyomi/AsagaoMix", + "downloads": 106, + "description": "📄 ライセンス / License修正 CreativeML OpenRAIL-M ライセンス / Modified CreativeML OpenRAIL-M licenseこのモデルのクレジットを入れずに使用するUse the model without crediting the creatorこのモデルで生成した画像を商用利用するSell images they generateこのモデルを商用の画像生成サービスで利用するRun on services that generate images for moneyこのモデルを使用したマージモデルを共有するShare merges using this modelこのモデル、またはこのモデルをマージしたモデルを販売するSell this model or merges using this modelこのモデルをマージしたモデルに異なる権限を設定するHave different permissions when sharing merges", "source": "Hugging Face", - "score": 1.4036259967655454e-05, - "project_name": "AfterRealXL" + "score": 1.5859920395169366e-05, + "project_name": "AsagaoMix" }, { - "url": "https://huggingface.co/Miwa-Keita/zenz-v1", + "url": "https://huggingface.co/nlp-waseda/roberta-large-japanese", + "downloads": 103, + "description": "nlp-waseda/roberta-large-japaneseModel descriptionThis is a Japanese RoBERTa large model pretrained on Japanese Wikipedia and the Japanese portion of CC-100.How to useYou can use this model for masked language modeling as follows:from transformers import AutoTokenizer, AutoModelForMaskedLMtokenizer = AutoTokenizer.from_pretrained(\"nlp-waseda/roberta-large-japanese\")", + "source": "Hugging Face", + "score": 1.541105472360797e-05, + "project_name": "roberta-large-japanese" + }, + { + "url": "https://huggingface.co/OrionStarAI/Orion-14B-LongChat", "downloads": 100, - "description": "zenz-v1zenz-v1はGPT-2アーキテクチャに基づくかな漢字変換タスクに特化した言語モデルです。", + "description": "Orion-14B🌐English | 🇨🇳中文 | 🇯🇵日本語 | 🇰🇷한국어🤗", "source": "Hugging Face", - "score": 1.3897287096688568e-05, - "project_name": "zenz-v1" + "score": 1.4962189052046573e-05, + "project_name": "Orion-14B-LongChat" }, { - "url": "https://huggingface.co/pfnet/plamo-13b-instruct-nc", - "downloads": 98, - "description": "PLaMo-13B-Instruct-NCModel DescriptionPLaMo-13B-Instruct-NC is a noncommercial instruct fine-tuned model built upon the 8192 context length version of PLaMo-13B text generation model.", + "url": "https://huggingface.co/TFMC/ChatNTQ-JA-7b-v1.0-GGUF", + "downloads": 97, + "description": "GGUF conversion of NTQAI/chatntq-ja-7b-v1.0ChatNTQ-JA-7b-v1.0 is a Japanese chat fine-tuned model built on top of the stabilityai/japanese-stablelm-base-gamma-7b, which is originally based on Mistral 7B v0.1.", "source": "Hugging Face", - "score": 1.3619341354754797e-05, - "project_name": "plamo-13b-instruct-nc" + "score": 1.4513323380485176e-05, + "project_name": "ChatNTQ-JA-7b-v1.0-GGUF" }, { - "url": "https://huggingface.co/owner203/japanese-llama-2-13b-gguf", - "downloads": 98, - "description": "Japanese-LLaMA-2-13B-GGUFJapanese-LLaMA-2-13B-GGUFはJapanese-LLaMA-2-13BのGGUF形式です。", + "url": "https://huggingface.co/stabilityai/japanese-stablelm-instruct-alpha-7b", + "downloads": 95, + "description": "This repository is publicly accessible, but you have to accept the conditions to access its files and content.", "source": "Hugging Face", - "score": 1.3619341354754797e-05, - "project_name": "japanese-llama-2-13b-gguf" + "score": 1.4214079599444245e-05, + "project_name": "japanese-stablelm-instruct-alpha-7b" }, { - "url": "https://huggingface.co/nlp-waseda/roberta-large-japanese-with-auto-jumanpp", - "downloads": 97, - "description": "nlp-waseda/roberta-large-japanese-with-auto-jumanppModel descriptionThis is a Japanese RoBERTa large model pretrained on Japanese Wikipedia and the Japanese portion of CC-100.How to useYou can use this model for masked language modeling as follows:from transformers import AutoTokenizer, AutoModelForMaskedLMtokenizer = AutoTokenizer.from_pretrained(\"nlp-waseda/roberta-large-japanese-with-auto-jumanpp\")", + "url": "https://huggingface.co/stabilityai/japanese-stablelm-base-ja_vocab-beta-7b", + "downloads": 95, + "description": "Japanese-StableLM-Base-JAVocab-Beta-7BA cute robot wearing a kimono writes calligraphy with one single brush — Stable Diffusion XLModel Descriptionjapanese-stablelm-base-ja_vocab-beta-7b is a 7B-parameter decoder-only language model based on Llama-2-7b that has been fine-tuned on a diverse collection of Japanese data, with the intent of maximizing downstream performance on Japanese language tasks.", "source": "Hugging Face", - "score": 1.348036848378791e-05, - "project_name": "roberta-large-japanese-with-auto-jumanpp" + "score": 1.4214079599444245e-05, + "project_name": "japanese-stablelm-base-ja_vocab-beta-7b" }, { - "url": "https://huggingface.co/sonoisa/sentence-t5-base-ja-mean-tokens", - "downloads": 97, - "description": "This is a Japanese sentence-T5 model.", + "url": "https://huggingface.co/studio-ousia/luke-japanese-large-lite", + "downloads": 93, + "description": "luke-japanese-large-liteluke-japanese is the Japanese version of LUKE (LanguageUnderstanding with Knowledge-based Embeddings), a pre-trainedknowledge-enhanced contextualized representation of words and entities.", "source": "Hugging Face", - "score": 1.348036848378791e-05, - "project_name": "sentence-t5-base-ja-mean-tokens" + "score": 1.3914835818403313e-05, + "project_name": "luke-japanese-large-lite" }, { - "url": "https://huggingface.co/TheBloke/japanese-stablelm-instruct-beta-70B-GPTQ", - "downloads": 94, - "description": "Chat & support: TheBloke's Discord serverWant to contribute?", + "url": "https://huggingface.co/nlp-waseda/bigbird-base-japanese", + "downloads": 92, + "description": "nlp-waseda/bigbird-base-japaneseModel descriptionThis is a Japanese BigBird base model pretrained on Japanese Wikipedia, the Japanese portion of CC-100, and the Japanese portion of OSCAR.How to useYou can use this model for masked language modeling as follows:from transformers import AutoTokenizer, AutoModelForMaskedLMtokenizer = AutoTokenizer.from_pretrained(\"nlp-waseda/bigbird-base-japanese\")", "source": "Hugging Face", - "score": 1.3063449870887253e-05, - "project_name": "japanese-stablelm-instruct-beta-70B-GPTQ" + "score": 1.3765213927882847e-05, + "project_name": "bigbird-base-japanese" }, { - "url": "https://huggingface.co/nlp-waseda/comet-gpt2-small-japanese", - "downloads": 94, - "description": "COMET-GPT2 jaFinetuned GPT-2 on ATOMIC ja using a causal language modeling (CLM) objective.", + "url": "https://huggingface.co/HODACHI/mistral-seven-merged", + "downloads": 90, + "description": "概要「HODACHI/mistral-seven-merged」は、Mistral系の高い日本語性能を誇る、７つのモデルを再トレーニングを行いながら評価を行い、マージした結果に基づき生まれたモデルです。", "source": "Hugging Face", - "score": 1.3063449870887253e-05, - "project_name": "comet-gpt2-small-japanese" + "score": 1.3465970146841915e-05, + "project_name": "mistral-seven-merged" }, { - "url": "https://huggingface.co/ku-nlp/bart-large-japanese", - "downloads": 93, - "description": "Model Card for Japanese BART largeModel descriptionThis is a Japanese BART large model pre-trained on Japanese Wikipedia.", + "url": "https://huggingface.co/stabilityai/japanese-instructblip-alpha", + "downloads": 90, + "description": "Japanese InstructBLIP AlphaModel DetailsJapanese InstructBLIP Alpha is a vision-language instruction-following model that enables to generate Japanese descriptions for input images and optionally input texts such as questions.", "source": "Hugging Face", - "score": 1.2924476999920369e-05, - "project_name": "bart-large-japanese" + "score": 1.3465970146841915e-05, + "project_name": "japanese-instructblip-alpha" }, { - "url": "https://huggingface.co/owner203/japanese-alpaca-2-13b-gguf", - "downloads": 91, - "description": "Japanese-Alpaca-2-13B-GGUFJapanese-Alpaca-2-13B-GGUFはJapanese-Alpaca-2-13BのGGUF形式です。", + "url": "https://huggingface.co/colorfulscoop/bert-base-ja", + "downloads": 89, + "description": "BERT base Japanese modelThis repository contains a BERT base model trained on Japanese Wikipedia dataset.", "source": "Hugging Face", - "score": 1.2646531257986598e-05, - "project_name": "japanese-alpaca-2-13b-gguf" + "score": 1.331634825632145e-05, + "project_name": "bert-base-ja" }, { - "url": "https://huggingface.co/zh-plus/faster-whisper-large-v2-japanese-5k-steps", - "downloads": 89, - "description": "Converted from clu-ling/whisper-large-v2-japanese-5k-steps using CTranslate2.Usage:Install pip install faster-whisper (Check faster-whisper for detailed instructions.", + "url": "https://huggingface.co/sonoisa/t5-qiita-title-generation", + "downloads": 88, + "description": "記事本文からタイトルを生成するモデルSEE: https://qiita.com/sonoisa/items/30876467ad5a8a81821f", "source": "Hugging Face", - "score": 1.2368585516052825e-05, - "project_name": "faster-whisper-large-v2-japanese-5k-steps" + "score": 1.3166726365800985e-05, + "project_name": "t5-qiita-title-generation" }, { - "url": "https://huggingface.co/tokyotech-llm/Swallow-7b-NVE-hf", - "downloads": 83, - "description": "SwallowOur Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data.", + "url": "https://huggingface.co/sazyou-roukaku/AfterRealXL", + "downloads": 85, + "description": "こちらでアップロードできないので、civitaiにて先に公開しています。", "source": "Hugging Face", - "score": 1.1534748290251512e-05, - "project_name": "Swallow-7b-NVE-hf" + "score": 1.2717860694239587e-05, + "project_name": "AfterRealXL" }, { - "url": "https://huggingface.co/stabilityai/japanese-stablelm-base-ja_vocab-beta-7b", - "downloads": 81, - "description": "Japanese-StableLM-Base-JAVocab-Beta-7BA cute robot wearing a kimono writes calligraphy with one single brush — Stable Diffusion XLModel Descriptionjapanese-stablelm-base-ja_vocab-beta-7b is a 7B-parameter decoder-only language model based on Llama-2-7b that has been fine-tuned on a diverse collection of Japanese data, with the intent of maximizing downstream performance on Japanese language tasks.", + "url": "https://huggingface.co/Miwa-Keita/zenz-v1", + "downloads": 85, + "description": "zenz-v1zenz-v1はGPT-2アーキテクチャに基づくかな漢字変換タスクに特化した言語モデルです。", "source": "Hugging Face", - "score": 1.125680254831774e-05, - "project_name": "japanese-stablelm-base-ja_vocab-beta-7b" + "score": 1.2717860694239587e-05, + "project_name": "zenz-v1" }, { - "url": "https://huggingface.co/sonoisa/t5-qiita-title-generation", - "downloads": 80, - "description": "記事本文からタイトルを生成するモデルSEE: https://qiita.com/sonoisa/items/30876467ad5a8a81821f", + "url": "https://huggingface.co/zh-plus/faster-whisper-large-v2-japanese-5k-steps", + "downloads": 85, + "description": "Converted from clu-ling/whisper-large-v2-japanese-5k-steps using CTranslate2.Usage:Install pip install faster-whisper (Check faster-whisper for detailed instructions.", "source": "Hugging Face", - "score": 1.1117829677350855e-05, - "project_name": "t5-qiita-title-generation" + "score": 1.2717860694239587e-05, + "project_name": "faster-whisper-large-v2-japanese-5k-steps" }, { - "url": "https://huggingface.co/stabilityai/japanese-stablelm-instruct-alpha-7b", - "downloads": 78, - "description": "This repository is publicly accessible, but you have to accept the conditions to access its files and content.", + "url": "https://huggingface.co/LoneStriker/SambaLingo-Japanese-Chat-GGUF", + "downloads": 83, + "description": "SambaLingo-Japanese-ChatSambaLingo-Japanese-Chat is a human aligned chat model trained in Japanese and English.", "source": "Hugging Face", - "score": 1.0839883935417084e-05, - "project_name": "japanese-stablelm-instruct-alpha-7b" + "score": 1.2418616913198655e-05, + "project_name": "SambaLingo-Japanese-Chat-GGUF" }, { - "url": "https://huggingface.co/rinna/nekomata-14b-gguf", - "downloads": 77, - "description": "rinna/nekomata-14b-ggufOverviewThe model is the GGUF version of rinna/nekomata-14b.", + "url": "https://huggingface.co/turing-motors/heron-chat-blip-ja-stablelm-base-7b-v1", + "downloads": 83, + "description": "Heron BLIP Japanese StableLM", "source": "Hugging Face", - "score": 1.0700911064450198e-05, - "project_name": "nekomata-14b-gguf" + "score": 1.2418616913198655e-05, + "project_name": "heron-chat-blip-ja-stablelm-base-7b-v1" }, { - "url": "https://huggingface.co/rinna/nekomata-7b-instruction-gguf", - "downloads": 77, - "description": "rinna/nekomata-7b-instruction-ggufOverviewThe model is the GGUF version of rinna/nekomata-7b-instruction.", + "url": "https://huggingface.co/ThePioneer/MoeDiffusionPlusPlus", + "downloads": 82, + "description": "モデル説明 (model explanation)V1 = MoeDiffusion 1.0 + (HassanBlend 1.5 - VMix03) * 0.2V2 = MoeDiffusion 0.6 : HassanBlend 1.5 0.2 : VMix03 : 0.2マージ元のルーツにNAIリークやInsta系モデルが含まれるという噂があるので、NAIリークアンチ・Insta系モデルアンチには非推奨理想の黒髪ポニテ顔が出せるYaguruMagikuを、ある程度顔が近くて制御しやすいAbyssOrangeMix2と混ぜてみた。", "source": "Hugging Face", - "score": 1.0700911064450198e-05, - "project_name": "nekomata-7b-instruction-gguf" + "score": 1.226899502267819e-05, + "project_name": "MoeDiffusionPlusPlus" }, { - "url": "https://huggingface.co/rinna/nekomata-7b-gguf", - "downloads": 76, - "description": "rinna/nekomata-7b-ggufOverviewThe model is the GGUF version of rinna/nekomata-7b.", + "url": "https://huggingface.co/tokyotech-llm/Swallow-13b-NVE-hf", + "downloads": 82, + "description": "SwallowOur Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data.", "source": "Hugging Face", - "score": 1.0561938193483311e-05, - "project_name": "nekomata-7b-gguf" + "score": 1.226899502267819e-05, + "project_name": "Swallow-13b-NVE-hf" }, { - "url": "https://huggingface.co/colorfulscoop/bert-base-ja", - "downloads": 75, - "description": "BERT base Japanese modelThis repository contains a BERT base model trained on Japanese Wikipedia dataset.", + "url": "https://huggingface.co/owner203/japanese-llama-3-8b-instruct-v2-gguf", + "downloads": 81, + "description": "Japanese-LLaMA-3-8B-Instruct-v2-GGUFJapanese-LLaMA-3-8B-Instruct-v2-GGUFはJapanese-LLaMA-3-8B-Instruct-v2のGGUF形式です。", "source": "Hugging Face", - "score": 1.0422965322516427e-05, - "project_name": "bert-base-ja" + "score": 1.2119373132157725e-05, + "project_name": "japanese-llama-3-8b-instruct-v2-gguf" }, { - "url": "https://huggingface.co/okazaki-lab/japanese-gpt2-medium-unidic", - "downloads": 75, - "description": "japanese-gpt2-medium-unidicThis is a medium-sized Japanese GPT-2 model using BERT-like tokenizer.", + "url": "https://huggingface.co/youhansun/Llama-3-70B-japanese-suzume-vector-v0.1-Q2_K-GGUF", + "downloads": 79, + "description": "youhansun/Llama-3-70B-japanese-suzume-vector-v0.1-Q2_K-GGUFThis model was converted to GGUF format from mmnga/Llama-3-70B-japanese-suzume-vector-v0.1 using llama.cpp via the ggml.ai's GGUF-my-repo space.", "source": "Hugging Face", - "score": 1.0422965322516427e-05, - "project_name": "japanese-gpt2-medium-unidic" + "score": 1.1820129351116793e-05, + "project_name": "Llama-3-70B-japanese-suzume-vector-v0.1-Q2_K-GGUF" }, { - "url": "https://huggingface.co/KoichiYasuoka/bert-base-japanese-char-extended", - "downloads": 75, - "description": "bert-base-japanese-char-extendedModel", + "url": "https://huggingface.co/owner203/japanese-llama-2-13b-gguf", + "downloads": 79, + "description": "Japanese-LLaMA-2-13B-GGUFJapanese-LLaMA-2-13B-GGUFはJapanese-LLaMA-2-13BのGGUF形式です。", "source": "Hugging Face", - "score": 1.0422965322516427e-05, - "project_name": "bert-base-japanese-char-extended" + "score": 1.1820129351116793e-05, + "project_name": "japanese-llama-2-13b-gguf" }, { "url": "https://huggingface.co/Ivydata/whisper-small-japanese", - "downloads": 71, + "downloads": 76, "description": "Fine-tuned Japanese Whisper model for speech recognition using whisper-smallFine-tuned openai/whisper-small on Japanese using Common Voice, JVS and JSUT.When using this model, make sure that your speech input is sampled at 16kHz.", "source": "Hugging Face", - "score": 9.867073838648883e-06, + "score": 1.1371263679555395e-05, "project_name": "whisper-small-japanese" }, { "url": "https://huggingface.co/Mizuiro-sakura/luke-japanese-base-marcja", - "downloads": 69, + "downloads": 76, "description": "このモデルはluke-japanese-baseをファインチューニングして、MARC-ja(positive or negativeの二値分類)に用いれるようにしたものです。", "source": "Hugging Face", - "score": 9.589128096715112e-06, + "score": 1.1371263679555395e-05, "project_name": "luke-japanese-base-marcja" }, { - "url": "https://huggingface.co/recruit-jp/japanese-clip-vit-b-32-roberta-base", - "downloads": 66, - "description": "recruit-jp/japanese-clip-vit-b-32-roberta-baseOverviewDeveloped by: Recruit Co.", + "url": "https://huggingface.co/okazaki-lab/japanese-gpt2-medium-unidic", + "downloads": 74, + "description": "japanese-gpt2-medium-unidicThis is a medium-sized Japanese GPT-2 model using BERT-like tokenizer.", "source": "Hugging Face", - "score": 9.172209483814455e-06, - "project_name": "japanese-clip-vit-b-32-roberta-base" + "score": 1.1072019898514463e-05, + "project_name": "japanese-gpt2-medium-unidic" }, { - "url": "https://huggingface.co/TeamFnord/manga-ocr", - "downloads": 66, - "description": "Manga OCROptical character recognition for Japanese text, with the main focus being Japanese manga.", + "url": "https://huggingface.co/umiyuki/Japanese-Chat-Umievo-itr004-7b", + "downloads": 73, + "description": "japanese-chat-umievo-itr004-7bThis is a merge of pre-trained language models created using mergekit.", "source": "Hugging Face", - "score": 9.172209483814455e-06, - "project_name": "manga-ocr" + "score": 1.0922398007993999e-05, + "project_name": "Japanese-Chat-Umievo-itr004-7b" }, { - "url": "https://huggingface.co/vumichien/wav2vec2-large-xlsr-japanese", - "downloads": 66, - "description": "Wav2Vec2-Large-XLSR-53-JapaneseFine-tuned facebook/wav2vec2-large-xlsr-53 on Japanese using the Common Voice and Japanese speech corpus of Saruwatari-lab, University of Tokyo JSUT.When using this model, make sure that your speech input is sampled at 16kHz.", + "url": "https://huggingface.co/owner203/japanese-alpaca-2-13b-gguf", + "downloads": 73, + "description": "Japanese-Alpaca-2-13B-GGUFJapanese-Alpaca-2-13B-GGUFはJapanese-Alpaca-2-13BのGGUF形式です。", "source": "Hugging Face", - "score": 9.172209483814455e-06, - "project_name": "wav2vec2-large-xlsr-japanese" + "score": 1.0922398007993999e-05, + "project_name": "japanese-alpaca-2-13b-gguf" }, { - "url": "https://huggingface.co/ThePioneer/MoeDiffusion", - "downloads": 65, - "description": "モデル説明 (model explanation)YaguruMagiku 0.6 : AbyssOrangeMix2_sfw 0.4マージ元のルーツにNAIリークが含まれるという噂があるので、NAIリークアンチには非推奨理想の黒髪ポニテ顔が出せるYaguruMagikuを、ある程度顔が近くて制御しやすいAbyssOrangeMix2と混ぜてみた。", + "url": "https://huggingface.co/vumichien/wav2vec2-large-xlsr-japanese", + "downloads": 67, + "description": "Wav2Vec2-Large-XLSR-53-JapaneseFine-tuned facebook/wav2vec2-large-xlsr-53 on Japanese using the Common Voice and Japanese speech corpus of Saruwatari-lab, University of Tokyo JSUT.When using this model, make sure that your speech input is sampled at 16kHz.", "source": "Hugging Face", - "score": 9.033236612847569e-06, - "project_name": "MoeDiffusion" + "score": 1.0024666664871203e-05, + "project_name": "wav2vec2-large-xlsr-japanese" }, { - "url": "https://huggingface.co/oshizo/japanese-e5-mistral-1.9b", - "downloads": 65, - "description": "Model trained on 800,000 Japanese sentences after reducing oshizo/japanese-e5-mistral-7b_slerp to 8 layers.", + "url": "https://huggingface.co/rinna/nekomata-7b-instruction-gguf", + "downloads": 67, + "description": "rinna/nekomata-7b-instruction-ggufOverviewThe model is the GGUF version of rinna/nekomata-7b-instruction.", "source": "Hugging Face", - "score": 9.033236612847569e-06, - "project_name": "japanese-e5-mistral-1.9b" + "score": 1.0024666664871203e-05, + "project_name": "nekomata-7b-instruction-gguf" }, { - "url": "https://huggingface.co/youhansun/Llama-3-70B-japanese-suzume-vector-v0.1-Q2_K-GGUF", - "downloads": 65, - "description": "youhansun/Llama-3-70B-japanese-suzume-vector-v0.1-Q2_K-GGUFThis model was converted to GGUF format from mmnga/Llama-3-70B-japanese-suzume-vector-v0.1 using llama.cpp via the ggml.ai's GGUF-my-repo space.", + "url": "https://huggingface.co/Local-Novel-LLM-project/Vecteus-V2-7B", + "downloads": 67, + "description": "Vecteus-V2-7Bこのモデルは、ベクトルマージなどを用い作成された高性能ベースモデルです。 ", "source": "Hugging Face", - "score": 9.033236612847569e-06, - "project_name": "Llama-3-70B-japanese-suzume-vector-v0.1-Q2_K-GGUF" + "score": 1.0024666664871203e-05, + "project_name": "Vecteus-V2-7B" }, { - "url": "https://huggingface.co/studio-ousia/luke-japanese-large-lite", - "downloads": 64, - "description": "luke-japanese-large-liteluke-japanese is the Japanese version of LUKE (LanguageUnderstanding with Knowledge-based Embeddings), a pre-trainedknowledge-enhanced contextualized representation of words and entities.", + "url": "https://huggingface.co/natsusakiyomi/Riga_Collection", + "downloads": 66, + "description": "Riga_collectionとは？", "source": "Hugging Face", - "score": 8.894263741880684e-06, - "project_name": "luke-japanese-large-lite" + "score": 9.875044774350739e-06, + "project_name": "Riga_Collection" }, { - "url": "https://huggingface.co/ysakuramoto/mobilebert-ja", - "downloads": 64, - "description": "MobileBERT 日本語事前学習済みモデル爆誕！！", + "url": "https://huggingface.co/votepurchase/Yaki-Dofu-Mix", + "downloads": 65, + "description": "Yaki-Dofu-Mix概要 / OverviewYaki-Dofu-Mixは、アニメ風の画風に特化したマージモデルです。 ", "source": "Hugging Face", - "score": 8.894263741880684e-06, - "project_name": "mobilebert-ja" + "score": 9.725422883830273e-06, + "project_name": "Yaki-Dofu-Mix" }, { - "url": "https://huggingface.co/Mizuiro-sakura/deberta-v2-base-japanese-finetuned-QAe", - "downloads": 61, - "description": "このモデルはdeberta-v2-base-japaneseをファインチューニングしてQAタスクに用いれるようにしたものです。", + "url": "https://huggingface.co/haqishen/h2o-Llama-3-8B-Japanese-Instruct", + "downloads": 63, + "description": "IntroductionWho am I: Qishen Ha", "source": "Hugging Face", - "score": 8.477345128980027e-06, - "project_name": "deberta-v2-base-japanese-finetuned-QAe" + "score": 9.426179102789341e-06, + "project_name": "h2o-Llama-3-8B-Japanese-Instruct" }, { - "url": "https://huggingface.co/ThePioneer/MoeSharpV1", - "downloads": 61, - "description": "モデル説明 (model explanation)MoeDiffusionPlusPlus 0.7 : DreamShaper 3.3 (full) 0.3。", + "url": "https://huggingface.co/rinna/nekomata-14b-gguf", + "downloads": 63, + "description": "rinna/nekomata-14b-ggufOverviewThe model is the GGUF version of rinna/nekomata-14b.", "source": "Hugging Face", - "score": 8.477345128980027e-06, - "project_name": "MoeSharpV1" + "score": 9.426179102789341e-06, + "project_name": "nekomata-14b-gguf" }, { - "url": "https://huggingface.co/thefrigidliquidation/nllb-jaen-1.3B-lightnovels", - "downloads": 60, - "description": "NLLB 1.3B fine-tuned on Japanese to English Light Novel translationThis model was fine-tuned on light and web novel for Japanese to English translation.", + "url": "https://huggingface.co/ku-nlp/gpt2-large-japanese-char", + "downloads": 63, + "description": "Model Card for Japanese character-level GPT-2 LargeModel descriptionThis is a Japanese character-level GPT-2 Large (717M parameters) language model pre-trained on Japanese Wikipedia, the Japanese portion of CC-100, and the Japanese portion of OSCAR.How to useYou can use this model directly with a pipeline for text generation.", "source": "Hugging Face", - "score": 8.33837225801314e-06, - "project_name": "nllb-jaen-1.3B-lightnovels" + "score": 9.426179102789341e-06, + "project_name": "gpt2-large-japanese-char" }, { - "url": "https://huggingface.co/Aratako/ELYZA-japanese-Llama-2-MoE-2x13B-v0.1-GGUF", - "downloads": 60, - "description": "ELYZA-japanese-Llama-2-MoE-2x13B-v0.1-GGUF概要Aratako/ELYZA-japanese-Llama-2-MoE-2x13B-v0.1の量子化済みGGUF版です。", + "url": "https://huggingface.co/TeamFnord/manga-ocr", + "downloads": 63, + "description": "Manga OCROptical character recognition for Japanese text, with the main focus being Japanese manga.", "source": "Hugging Face", - "score": 8.33837225801314e-06, - "project_name": "ELYZA-japanese-Llama-2-MoE-2x13B-v0.1-GGUF" + "score": 9.426179102789341e-06, + "project_name": "manga-ocr" }, { - "url": "https://huggingface.co/Lasorco/spekulatius", - "downloads": 60, - "description": "spekulatiusマージしているとたまに出てくる「目的の意図とは違うのだけどなんだか消すにはもったいないモデル」をおすそ分けするシリーズです。", + "url": "https://huggingface.co/Local-Novel-LLM-project/Ninja-v1", + "downloads": 61, + "description": "Our ModelsVecteusNinja-v1Ninja-v1-NSFWNinja-v1-128kNinja-v1-NSFW-128kModel Card for Ninja-v1.0The Mistral-7B--based Large Language Model (LLM) is an noveldataset fine-tuned version of the Mistral-7B-v0.1Ninja has the following changes compared to Mistral-7B-v0.1.Achieving both high quality Japanese and English generationMemory ability that does not forget even after long-context generationThis model was created with the help of GPUs from the first LocalAI hackathon.", "source": "Hugging Face", - "score": 8.33837225801314e-06, - "project_name": "spekulatius" + "score": 9.126935321748409e-06, + "project_name": "Ninja-v1" }, { "url": "https://huggingface.co/Mizuiro-sakura/luke-japanese-base-finetuned-jsts", - "downloads": 58, + "downloads": 59, "description": "このモデルはluke-japanese-baseをファインチューニングして、JSTS(文章の類似度計算)に用いれるようにしたものです。", "source": "Hugging Face", - "score": 8.06042651607937e-06, + "score": 8.827691540707479e-06, "project_name": "luke-japanese-base-finetuned-jsts" }, { - "url": "https://huggingface.co/haqishen/h2o-Llama-3-8B-Japanese-Instruct", - "downloads": 58, - "description": "IntroductionWho am I: Qishen Ha", + "url": "https://huggingface.co/nlp-waseda/gpt2-xl-japanese", + "downloads": 59, + "description": "nlp-waseda/gpt2-xl-japaneseThis is Japanese GPT2 with approximately　1.5B parameters pretrained on Japanese Wikipedia and CC-100The model architecture of the model are based on Radford+ 2019.Intended uses & limitationsYou can use the raw model for text generation or fine-tune it to a downstream task.", "source": "Hugging Face", - "score": 8.06042651607937e-06, - "project_name": "h2o-Llama-3-8B-Japanese-Instruct" + "score": 8.827691540707479e-06, + "project_name": "gpt2-xl-japanese" }, { - "url": "https://huggingface.co/KoichiYasuoka/bert-base-japanese-unidic-luw-upos", - "downloads": 57, - "description": "bert-base-japanese-unidic-luw-uposModel", + "url": "https://huggingface.co/rinna/nekomata-7b-gguf", + "downloads": 58, + "description": "rinna/nekomata-7b-ggufOverviewThe model is the GGUF version of rinna/nekomata-7b.", "source": "Hugging Face", - "score": 7.921453645112483e-06, - "project_name": "bert-base-japanese-unidic-luw-upos" + "score": 8.678069650187013e-06, + "project_name": "nekomata-7b-gguf" }, { - "url": "https://huggingface.co/turing-motors/heron-chat-blip-ja-stablelm-base-7b-v0", - "downloads": 56, - "description": "Heron BLIP Japanese StableLM", + "url": "https://huggingface.co/ku-nlp/bart-large-japanese", + "downloads": 57, + "description": "Model Card for Japanese BART largeModel descriptionThis is a Japanese BART large model pre-trained on Japanese Wikipedia.", "source": "Hugging Face", - "score": 7.782480774145599e-06, - "project_name": "heron-chat-blip-ja-stablelm-base-7b-v0" + "score": 8.528447759666547e-06, + "project_name": "bart-large-japanese" }, { - "url": "https://huggingface.co/nlp-waseda/roberta_jtruthfulqa", - "downloads": 55, - "description": "Finetuned Waseda RoBERTa to evaluate the generated answers on JTruthfulQA.", + "url": "https://huggingface.co/Mizuiro-sakura/deberta-v2-base-japanese-finetuned-QAe", + "downloads": 57, + "description": "このモデルはdeberta-v2-base-japaneseをファインチューニングしてQAタスクに用いれるようにしたものです。", "source": "Hugging Face", - "score": 7.643507903178712e-06, - "project_name": "roberta_jtruthfulqa" + "score": 8.528447759666547e-06, + "project_name": "deberta-v2-base-japanese-finetuned-QAe" }, { - "url": "https://huggingface.co/Lasorco/Kokuwa", - "downloads": 55, - "description": "Kokuwalamettaの改良でマージさせるモデル探しをしていたらKiwiMixという面白そうなモデルを見つけました。", + "url": "https://huggingface.co/stabilityai/japanese-stablelm-instruct-beta-7b", + "downloads": 56, + "description": "Japanese-StableLM-Instruct-Beta-7BA cute robot wearing a kimono writes calligraphy with one single brush — Stable Diffusion XLModel Descriptionjapanese-stablelm-instruct-beta-7b is a 7B-parameter decoder-only language model based on", "source": "Hugging Face", - "score": 7.643507903178712e-06, - "project_name": "Kokuwa" + "score": 8.378825869146081e-06, + "project_name": "japanese-stablelm-instruct-beta-7b" }, { - "url": "https://huggingface.co/nlp-waseda/gpt2-xl-japanese", - "downloads": 54, - "description": "nlp-waseda/gpt2-xl-japaneseThis is Japanese GPT2 with approximately　1.5B parameters pretrained on Japanese Wikipedia and CC-100The model architecture of the model are based on Radford+ 2019.Intended uses & limitationsYou can use the raw model for text generation or fine-tune it to a downstream task.", + "url": "https://huggingface.co/turing-motors/heron-chat-blip-ja-stablelm-base-7b-v0", + "downloads": 55, + "description": "Heron BLIP Japanese StableLM", "source": "Hugging Face", - "score": 7.504535032211827e-06, - "project_name": "gpt2-xl-japanese" + "score": 8.229203978625615e-06, + "project_name": "heron-chat-blip-ja-stablelm-base-7b-v0" }, { - "url": "https://huggingface.co/Helsinki-NLP/opus-mt-ja-pl", - "downloads": 54, - "description": "jpn-polsource group: Japanesetarget group: PolishOPUS readme: jpn-polmodel: transformer-alignsource language(s): jpn jpn_Bopo jpn_Hani jpn_Hira jpn_Kana jpn_Latntarget language(s): polmodel: transformer-alignpre-processing: normalization + SentencePiece (spm32k,spm32k)", + "url": "https://huggingface.co/sonoisa/vl-t5-base-japanese", + "downloads": 55, + "description": "日本語VL-T5事前学習済みモデルThis is a VL-T5 (Unifying Vision-and-Language Tasks via Text Generation) model pretrained on Japanese corpus.", "source": "Hugging Face", - "score": 7.504535032211827e-06, - "project_name": "opus-mt-ja-pl" + "score": 8.229203978625615e-06, + "project_name": "vl-t5-base-japanese" }, { - "url": "https://huggingface.co/aixsatoshi/Swallow-MX-8x7b-NVE-chatvector-Mixtral-instruct-v2", + "url": "https://huggingface.co/aerner/lm-v2", "downloads": 54, - "description": "Swallow-MX-8x7b-NVE-v0.1に対し、Mixtral-8x7B-Instruct-v0.1とMixtral-8x7B-v0.1の差分をマージしたモデルです。", + "description": "Aerner LM-v2事前学習から全部日本語で学習させたモデルのバージョン2です。", "source": "Hugging Face", - "score": 7.504535032211827e-06, - "project_name": "Swallow-MX-8x7b-NVE-chatvector-Mixtral-instruct-v2" + "score": 8.079582088105149e-06, + "project_name": "lm-v2" }, { - "url": "https://huggingface.co/sonoisa/t5-base-japanese-adapt", - "downloads": 53, - "description": "日本語T5 Prefix Language ModelThis is a T5 (Text-to-Text Transfer Transformer)", + "url": "https://huggingface.co/slplab/wav2vec2-xls-r-300m-japanese-hiragana", + "downloads": 52, + "description": "Wav2Vec2-XLS-R-300M-Japanese-HiraganaFine-tuned facebook/wav2vec2-xls-r-300m on Japanese Hiragana characters using the Common Voice and JSUT.The sentence outputs do not contain word boundaries.", "source": "Hugging Face", - "score": 7.365562161244941e-06, - "project_name": "t5-base-japanese-adapt" + "score": 7.780338307064219e-06, + "project_name": "wav2vec2-xls-r-300m-japanese-hiragana" }, { - "url": "https://huggingface.co/slplab/wav2vec2-xls-r-300m-japanese-hiragana", - "downloads": 53, - "description": "Wav2Vec2-XLS-R-300M-Japanese-HiraganaFine-tuned facebook/wav2vec2-xls-r-300m on Japanese Hiragana characters using the Common Voice and JSUT.The sentence outputs do not contain word boundaries.", + "url": "https://huggingface.co/sonoisa/t5-base-japanese-mC4-Wikipedia", + "downloads": 52, + "description": "日本語T5事前学習済みモデルThis is a T5 (Text-to-Text Transfer Transformer) model pretrained on Japanese corpus.", "source": "Hugging Face", - "score": 7.365562161244941e-06, - "project_name": "wav2vec2-xls-r-300m-japanese-hiragana" + "score": 7.780338307064219e-06, + "project_name": "t5-base-japanese-mC4-Wikipedia" }, { - "url": "https://huggingface.co/ku-nlp/gpt2-large-japanese-char", + "url": "https://huggingface.co/Lasorco/spekulatius", "downloads": 51, - "description": "Model Card for Japanese character-level GPT-2 LargeModel descriptionThis is a Japanese character-level GPT-2 Large (717M parameters) language model pre-trained on Japanese Wikipedia, the Japanese portion of CC-100, and the Japanese portion of OSCAR.How to useYou can use this model directly with a pipeline for text generation.", + "description": "spekulatiusマージしているとたまに出てくる「目的の意図とは違うのだけどなんだか消すにはもったいないモデル」をおすそ分けするシリーズです。", "source": "Hugging Face", - "score": 7.0876164193111695e-06, - "project_name": "gpt2-large-japanese-char" + "score": 7.630716416543753e-06, + "project_name": "spekulatius" }, { - "url": "https://huggingface.co/tokyotech-llm/Swallow-13b-NVE-hf", + "url": "https://huggingface.co/llm-book/bert-base-japanese-v3-bpr-question-aio", "downloads": 50, - "description": "SwallowOur Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data.", + "description": "bert-base-japanese-v3-bpr-question-aio「大規模言語モデル入門」の第9章で紹介している文書検索モデルBPRの質問エンコーダです。", "source": "Hugging Face", - "score": 6.948643548344284e-06, - "project_name": "Swallow-13b-NVE-hf" + "score": 7.481094526023287e-06, + "project_name": "bert-base-japanese-v3-bpr-question-aio" }, { - "url": "https://huggingface.co/aerner/lm-v2", + "url": "https://huggingface.co/ThePioneer/MoeDiffusion", "downloads": 49, - "description": "Aerner LM-v2事前学習から全部日本語で学習させたモデルのバージョン2です。", + "description": "モデル説明 (model explanation)YaguruMagiku 0.6 : AbyssOrangeMix2_sfw 0.4マージ元のルーツにNAIリークが含まれるという噂があるので、NAIリークアンチには非推奨理想の黒髪ポニテ顔が出せるYaguruMagikuを、ある程度顔が近くて制御しやすいAbyssOrangeMix2と混ぜてみた。", "source": "Hugging Face", - "score": 6.8096706773773985e-06, - "project_name": "lm-v2" + "score": 7.331472635502821e-06, + "project_name": "MoeDiffusion" }, { - "url": "https://huggingface.co/Aratako/ELYZA-japanese-Llama-2-MoE-2x7B-v0.1-GGUF", + "url": "https://huggingface.co/sappho192/aihub-ja-ko-translator", "downloads": 49, - "description": "ELYZA-japanese-Llama-2-MoE-2x7B-v0.1-GGUF概要Aratako/ELYZA-japanese-Llama-2-MoE-2x7B-v0.1の量子化済みGGUF版です。", + "description": "Japanese to Korean translatorJapanese to Korean translator model based on EncoderDecoderModel(bert-japanese+kogpt2)", "source": "Hugging Face", - "score": 6.8096706773773985e-06, - "project_name": "ELYZA-japanese-Llama-2-MoE-2x7B-v0.1-GGUF" + "score": 7.331472635502821e-06, + "project_name": "aihub-ja-ko-translator" }, { - "url": "https://huggingface.co/turing-motors/heron-chat-blip-ja-stablelm-base-7b-v1", - "downloads": 48, - "description": "Heron BLIP Japanese StableLM", + "url": "https://huggingface.co/thefrigidliquidation/nllb-jaen-1.3B-lightnovels", + "downloads": 49, + "description": "NLLB 1.3B fine-tuned on Japanese to English Light Novel translationThis model was fine-tuned on light and web novel for Japanese to English translation.", "source": "Hugging Face", - "score": 6.670697806410513e-06, - "project_name": "heron-chat-blip-ja-stablelm-base-7b-v1" + "score": 7.331472635502821e-06, + "project_name": "nllb-jaen-1.3B-lightnovels" }, { - "url": "https://huggingface.co/llm-book/bert-base-japanese-v3-jcommonsenseqa", + "url": "https://huggingface.co/tokyotech-llm/Swallow-7b-NVE-hf", "downloads": 48, - "description": "bert-base-japanese-v3-jcommonsenseqa「大規模言語モデル入門」の第5章で紹介している(多肢選択式質問応答)のモデルです。", + "description": "SwallowOur Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data.", "source": "Hugging Face", - "score": 6.670697806410513e-06, - "project_name": "bert-base-japanese-v3-jcommonsenseqa" + "score": 7.181850744982355e-06, + "project_name": "Swallow-7b-NVE-hf" }, { - "url": "https://huggingface.co/AndrewMcDowell/wav2vec2-xls-r-300m-japanese", + "url": "https://huggingface.co/ThePioneer/MoeSharpV1", "downloads": 48, - "description": "This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the MOZILLA-FOUNDATION/COMMON_VOICE_8_0 - JA dataset.", + "description": "モデル説明 (model explanation)MoeDiffusionPlusPlus 0.7 : DreamShaper 3.3 (full) 0.3。", "source": "Hugging Face", - "score": 6.670697806410513e-06, - "project_name": "wav2vec2-xls-r-300m-japanese" + "score": 7.181850744982355e-06, + "project_name": "MoeSharpV1" }, { - "url": "https://huggingface.co/fukugawa/transformer-lm-japanese-0.1b", - "downloads": 47, - "description": "transformer-lm-japanese-0.1bThis is a JAX/Flax-based transformer language model trained on a Japanese dataset.", + "url": "https://huggingface.co/Helsinki-NLP/opus-mt-ja-pl", + "downloads": 48, + "description": "jpn-polsource group: Japanesetarget group: PolishOPUS readme: jpn-polmodel: transformer-alignsource language(s): jpn jpn_Bopo jpn_Hani jpn_Hira jpn_Kana jpn_Latntarget language(s): polmodel: transformer-alignpre-processing: normalization + SentencePiece (spm32k,spm32k)", "source": "Hugging Face", - "score": 6.531724935443627e-06, - "project_name": "transformer-lm-japanese-0.1b" + "score": 7.181850744982355e-06, + "project_name": "opus-mt-ja-pl" }, { - "url": "https://huggingface.co/sappho192/aihub-ja-ko-translator", + "url": "https://huggingface.co/sonoisa/t5-base-japanese-adapt", "downloads": 47, - "description": "Japanese to Korean translatorJapanese to Korean translator model based on EncoderDecoderModel(bert-japanese+kogpt2)", + "description": "日本語T5 Prefix Language ModelThis is a T5 (Text-to-Text Transfer Transformer)", "source": "Hugging Face", - "score": 6.531724935443627e-06, - "project_name": "aihub-ja-ko-translator" + "score": 7.03222885446189e-06, + "project_name": "t5-base-japanese-adapt" }, { - "url": "https://huggingface.co/abhishek/autonlp-japanese-sentiment-59362", - "downloads": 45, - "description": "Model Trained Using AutoNLPProblem type: Binary ClassificationModel ID: 59362Validation MetricsLoss: 0.13092292845249176Accuracy: 0.9527127414314258Precision: 0.9634070704982427Recall: 0.9842171959602166AUC: 0.9667289746092403F1:", + "url": "https://huggingface.co/llm-book/bert-base-japanese-v3-jcommonsenseqa", + "downloads": 47, + "description": "bert-base-japanese-v3-jcommonsenseqa「大規模言語モデル入門」の第5章で紹介している(多肢選択式質問応答)のモデルです。", "source": "Hugging Face", - "score": 6.253779193509856e-06, - "project_name": "autonlp-japanese-sentiment-59362" + "score": 7.03222885446189e-06, + "project_name": "bert-base-japanese-v3-jcommonsenseqa" }, { - "url": "https://huggingface.co/stabilityai/japanese-stablelm-instruct-beta-7b", - "downloads": 45, - "description": "Japanese-StableLM-Instruct-Beta-7BA cute robot wearing a kimono writes calligraphy with one single brush — Stable Diffusion XLModel Descriptionjapanese-stablelm-instruct-beta-7b is a 7B-parameter decoder-only language model based on", + "url": "https://huggingface.co/Aratako/Ninja-v1-RP", + "downloads": 47, + "description": "Ninja-v1-RPGGUF版はこちら/Click here for the GGUF version概要This is a merge of pre-trained language models created using mergekit.Aratako/Ninja-v1-RP-WIPをベースに、Task Vectorの加算・Model Stockによるマージを行い指示追従能力と表現力を強化したロールプレイ用モデルです。", "source": "Hugging Face", - "score": 6.253779193509856e-06, - "project_name": "japanese-stablelm-instruct-beta-7b" + "score": 7.03222885446189e-06, + "project_name": "Ninja-v1-RP" }, { - "url": "https://huggingface.co/aixsatoshi/Swallow-MX-8x7b-NVE-chatvector-Mixtral-instruct", - "downloads": 44, - "description": "更新情報日本語機能とinstructベクトルのバランス調整したver.2をアップロードしましたSwallow-MX-8x7b-NVE-chatvector-Mixtral-instruct-v2モデル概要Swallow-MX-8x7b-NVE-v0.1に対し、Mixtral-8x7B-Instruct-v0.1とMixtral-8x7B-v0.1の差分をマージしたモデルです。", + "url": "https://huggingface.co/Lasorco/Kokuwa", + "downloads": 46, + "description": "Kokuwalamettaの改良でマージさせるモデル探しをしていたらKiwiMixという面白そうなモデルを見つけました。", "source": "Hugging Face", - "score": 6.11480632254297e-06, - "project_name": "Swallow-MX-8x7b-NVE-chatvector-Mixtral-instruct" + "score": 6.882606963941424e-06, + "project_name": "Kokuwa" }, { - "url": "https://huggingface.co/retrieva-jp/t5-small-long", - "downloads": 44, - "description": "Model card for model IDThis is a T5 v1.1 model, pre-trained on a Japanese corpus.", + "url": "https://huggingface.co/AndrewMcDowell/wav2vec2-xls-r-300m-japanese", + "downloads": 45, + "description": "This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the MOZILLA-FOUNDATION/COMMON_VOICE_8_0 - JA dataset.", "source": "Hugging Face", - "score": 6.11480632254297e-06, - "project_name": "t5-small-long" + "score": 6.732985073420958e-06, + "project_name": "wav2vec2-xls-r-300m-japanese" }, { - "url": "https://huggingface.co/sonoisa/sentence-bert-base-ja-en-mean-tokens", + "url": "https://huggingface.co/Aratako/ELYZA-japanese-Llama-2-MoE-2x13B-v0.1-GGUF", "downloads": 44, - "description": "This is a Japanese+English sentence-BERT model.", + "description": "ELYZA-japanese-Llama-2-MoE-2x13B-v0.1-GGUF概要Aratako/ELYZA-japanese-Llama-2-MoE-2x13B-v0.1の量子化済みGGUF版です。", "source": "Hugging Face", - "score": 6.11480632254297e-06, - "project_name": "sentence-bert-base-ja-en-mean-tokens" + "score": 6.583363182900493e-06, + "project_name": "ELYZA-japanese-Llama-2-MoE-2x13B-v0.1-GGUF" }, { - "url": "https://huggingface.co/kubota/luke-large-defamation-detection-japanese", - "downloads": 43, - "description": "luke-large-defamation-detection-japanese日本語誹謗中傷検出器This model is a fine-tuned version of studio-ousia/luke-japanese-large for the Japanese language finetuned for automatic defamation detection.", + "url": "https://huggingface.co/ysakuramoto/mobilebert-ja", + "downloads": 44, + "description": "MobileBERT 日本語事前学習済みモデル爆誕！！", "source": "Hugging Face", - "score": 5.975833451576084e-06, - "project_name": "luke-large-defamation-detection-japanese" + "score": 6.583363182900493e-06, + "project_name": "mobilebert-ja" }, { - "url": "https://huggingface.co/llm-book/bert-base-japanese-v3-bpr-question-aio", - "downloads": 43, - "description": "bert-base-japanese-v3-bpr-question-aio「大規模言語モデル入門」の第9章で紹介している文書検索モデルBPRの質問エンコーダです。", + "url": "https://huggingface.co/ken11/albert-base-japanese-v1-with-japanese-tokenizer", + "downloads": 44, + "description": "albert-base-japanese-v1-with-japanese日本語事前学習済みALBERTモデルですこのモデルではTokenizerにBertJapaneseTokenizerクラスを利用していますalbert-base-japanese-v1よりトークナイズ処理が楽になっていますHow to useファインチューニングこのモデルはPreTrainedモデルです基本的には各種タスク用にファインチューニングして使用されることを想定していますFill-Maskfor PyTorchfrom transformers import (AutoModelForMaskedLM, AutoTokenizer)tokenizer = AutoTokenizer.from_pretrained(\"ken11/albert-base-japanese-v1-with-japanese-tokenizer\")", "source": "Hugging Face", - "score": 5.975833451576084e-06, - "project_name": "bert-base-japanese-v3-bpr-question-aio" + "score": 6.583363182900493e-06, + "project_name": "albert-base-japanese-v1-with-japanese-tokenizer" }, { - "url": "https://huggingface.co/wolf4032/bert-japanese-token-classification-search-local-cuisine", + "url": "https://huggingface.co/Ivydata/whisper-base-japanese", "downloads": 43, - "description": "Model Card for Model ID料理を検索するための質問文から、検索検索用キーワードである固有表現を抽出しますModel DetailsModel Description例えば、「東京の肉料理で、春に食べられる、鶏肉を使った料理を教えてください」という文章を入力すると、「東京　→　都道府県/地方(AREA)」　「肉料理　→　種類(TYPE)」　「春　→　季節(SZN)」　「鶏肉　→　食材(INGR)」のように、固有表現を抽出します抽出対象は、AREA、TYPE、SZN、INGRの４つですLanguage(s) (NLP): 日本語License: mitFinetuned from model: tohoku-nlp/bert-base-japanese-v2Model SourcesRepository: wolf4032/nlp-token-classificationデータセット、言語モデル、アプリの作成に使ったコードが掲載されていますDocumentation: QiitaDemo: wolf4032/japanese-token-classification-search-local-", + "description": "Fine-tuned Japanese Whisper model for speech recognition using whisper-baseFine-tuned openai/whisper-base on Japanese using Common Voice, JVS and JSUT.When using this model, make sure that your speech input is sampled at 16kHz.", "source": "Hugging Face", - "score": 5.975833451576084e-06, - "project_name": "bert-japanese-token-classification-search-local-cuisine" + "score": 6.433741292380027e-06, + "project_name": "whisper-base-japanese" }, { - "url": "https://huggingface.co/cinmodel/electra-small-japanese-discriminator", - "downloads": 43, - "description": "Japanese ELECTRA-smallWe provide a Japanese ELECTRA-Small model, as described in ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators.", + "url": "https://huggingface.co/microsoft/unihanlm-base", + "downloads": 42, + "description": "Unihan LM: Coarse-to-Fine Chinese-Japanese Language Model Pretraining with the Unihan DatabaseModel descriptionChinese and Japanese share many characters with similar surface morphology.", "source": "Hugging Face", - "score": 5.975833451576084e-06, - "project_name": "electra-small-japanese-discriminator" + "score": 6.284119401859561e-06, + "project_name": "unihanlm-base" }, { "url": "https://huggingface.co/daisaku-s/medtxt_ner_roberta", - "downloads": 42, + "downloads": 41, "description": "日本語医療固有表現抽出モデル概要ソーシャル・コンピューティング研究室さまより公開されているMedTxt-CRを用いて、alabniiさまより公開されているRoBERTaをfine-tuningした固有表現抽出モデルです。", "source": "Hugging Face", - "score": 5.8368605806091985e-06, + "score": 6.134497511339095e-06, "project_name": "medtxt_ner_roberta" }, { - "url": "https://huggingface.co/Ivydata/whisper-base-japanese", - "downloads": 42, - "description": "Fine-tuned Japanese Whisper model for speech recognition using whisper-baseFine-tuned openai/whisper-base on Japanese using Common Voice, JVS and JSUT.When using this model, make sure that your speech input is sampled at 16kHz.", + "url": "https://huggingface.co/llm-book/bert-base-japanese-v3-bpr-passage-aio", + "downloads": 41, + "description": "bert-base-japanese-v3-bpr-passage-aio「大規模言語モデル入門」の第9章で紹介している文書検索モデルBPRのパッセージエンコーダです。", "source": "Hugging Face", - "score": 5.8368605806091985e-06, - "project_name": "whisper-base-japanese" + "score": 6.134497511339095e-06, + "project_name": "bert-base-japanese-v3-bpr-passage-aio" }, { - "url": "https://huggingface.co/ken11/albert-base-japanese-v1-with-japanese-tokenizer", + "url": "https://huggingface.co/retrieva-jp/t5-small-long", "downloads": 40, - "description": "albert-base-japanese-v1-with-japanese日本語事前学習済みALBERTモデルですこのモデルではTokenizerにBertJapaneseTokenizerクラスを利用していますalbert-base-japanese-v1よりトークナイズ処理が楽になっていますHow to useファインチューニングこのモデルはPreTrainedモデルです基本的には各種タスク用にファインチューニングして使用されることを想定していますFill-Maskfor PyTorchfrom transformers import (AutoModelForMaskedLM, AutoTokenizer)tokenizer = AutoTokenizer.from_pretrained(\"ken11/albert-base-japanese-v1-with-japanese-tokenizer\")", + "description": "Model card for model IDThis is a T5 v1.1 model, pre-trained on a Japanese corpus.", "source": "Hugging Face", - "score": 5.5589148386754275e-06, - "project_name": "albert-base-japanese-v1-with-japanese-tokenizer" + "score": 5.9848756208186295e-06, + "project_name": "t5-small-long" }, { - "url": "https://huggingface.co/Aratako/ELYZA-japanese-Llama-2-fast-MoE-2x7B-v0.1-GGUF", - "downloads": 38, - "description": "ELYZA-japanese-Llama-2-fast-MoE-2x7B-v0.1-GGUF概要Aratako/ELYZA-japanese-Llama-2-fast-MoE-2x7B-v0.1の量子化済みGGUF版です。", + "url": "https://huggingface.co/kubota/luke-large-defamation-detection-japanese", + "downloads": 40, + "description": "luke-large-defamation-detection-japanese日本語誹謗中傷検出器This model is a fine-tuned version of studio-ousia/luke-japanese-large for the Japanese language finetuned for automatic defamation detection.", "source": "Hugging Face", - "score": 5.280969096741656e-06, - "project_name": "ELYZA-japanese-Llama-2-fast-MoE-2x7B-v0.1-GGUF" + "score": 5.9848756208186295e-06, + "project_name": "luke-large-defamation-detection-japanese" }, { - "url": "https://huggingface.co/Aratako/Ninja-v1-RP", - "downloads": 38, - "description": "Ninja-v1-RPGGUF版はこちら/Click here for the GGUF version概要This is a merge of pre-trained language models created using mergekit.Aratako/Ninja-v1-RP-WIPをベースに、Task Vectorの加算・Model Stockによるマージを行い指示追従能力と表現力を強化したロールプレイ用モデルです。", + "url": "https://huggingface.co/KoichiYasuoka/bert-base-japanese-unidic-luw-upos", + "downloads": 40, + "description": "bert-base-japanese-unidic-luw-uposModel", "source": "Hugging Face", - "score": 5.280969096741656e-06, - "project_name": "Ninja-v1-RP" + "score": 5.9848756208186295e-06, + "project_name": "bert-base-japanese-unidic-luw-upos" }, { - "url": "https://huggingface.co/ku-accms/roberta-base-japanese-ssuw", - "downloads": 37, - "description": "ku-accms/roberta-base-japanese-ssuwModel descriptionThis is a pre-trained Japanese RoBERTa base model for super short unit words (SSUW).", + "url": "https://huggingface.co/tokyotech-llm/Swallow-70b-NVE-hf", + "downloads": 39, + "description": "SwallowOur Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data.", "source": "Hugging Face", - "score": 5.14199622577477e-06, - "project_name": "roberta-base-japanese-ssuw" + "score": 5.8352537302981636e-06, + "project_name": "Swallow-70b-NVE-hf" }, { - "url": "https://huggingface.co/oshizo/japanese-e5-mistral-7b_slerp", - "downloads": 37, - "description": "This model was created by merging intfloat/e5-mistral-7b-instruct and stabilityai/japanese-stablelm-base-gamma-7b.", + "url": "https://huggingface.co/bennexx/cl-tohoku-bert-base-japanese-v3-jlpt-classifier", + "downloads": 38, + "description": "SummaryThis is a text classifier for assigning a JLPT level.", "source": "Hugging Face", - "score": 5.14199622577477e-06, - "project_name": "japanese-e5-mistral-7b_slerp" + "score": 5.685631839777698e-06, + "project_name": "cl-tohoku-bert-base-japanese-v3-jlpt-classifier" }, { - "url": "https://huggingface.co/SkelterLabsInc/bert-base-japanese-jaquad", - "downloads": 37, - "description": "BERT base Japanese - JaQuADDescriptionA Japanese Question Answering model fine-tuned on JaQuAD.Please refer BERT base Japanese for details about the pre-training model.", + "url": "https://huggingface.co/shinyice/chatvector-llava-v1.5-plus-houou-v3-7b", + "downloads": 38, + "description": "Chatvector-llava-v1.5-plus-Houou-v3-7b Model CardModel Details※好奇心から生まれたモデルです。", "source": "Hugging Face", - "score": 5.14199622577477e-06, - "project_name": "bert-base-japanese-jaquad" + "score": 5.685631839777698e-06, + "project_name": "chatvector-llava-v1.5-plus-houou-v3-7b" + }, + { + "url": "https://huggingface.co/Mizuiro-sakura/t5-CAMERA-title-generation", + "downloads": 38, + "description": "sonoisa/t5-base-japaneseをファインチューニングして、タイトル生成に用いれるようにしたモデルです。", + "source": "Hugging Face", + "score": 5.685631839777698e-06, + "project_name": "t5-CAMERA-title-generation" }, { "url": "https://huggingface.co/umiyuki/Llama-3-Umievo-itr014-Shizuko-8b", "downloads": 37, "description": "Llama-3-Umievo-itr014-Shizuko-8bこのモデルは日本語に対応しているLlama-3ベースの４つのモデルを進化的アルゴリズムで進化的マージしたものです。", "source": "Hugging Face", - "score": 5.14199622577477e-06, + "score": 5.536009949257232e-06, "project_name": "Llama-3-Umievo-itr014-Shizuko-8b" }, { - "url": "https://huggingface.co/KoichiYasuoka/roberta-base-japanese-aozora", - "downloads": 36, - "description": "roberta-base-japanese-aozoraModel DescriptionThis is a RoBERTa model pre-trained on 青空文庫 texts with Japanese-LUW-Tokenizer.", + "url": "https://huggingface.co/rinna/japanese-stable-diffusion", + "downloads": 37, + "description": "One more step before getting this model.", "source": "Hugging Face", - "score": 5.003023354807885e-06, - "project_name": "roberta-base-japanese-aozora" + "score": 5.536009949257232e-06, + "project_name": "japanese-stable-diffusion" }, { - "url": "https://huggingface.co/abeja/Mixtral-8x7B-Instruct-v0.1-japanese", - "downloads": 36, - "description": "Mixtral-8x7B-Instruct-v0.1-japaneseMixtral-8x7B-Instruct-v0.1-japaneseはMixtral-8x7B-Instruct-v0.1をベースに日本語の語彙拡張継続事前学習を実施したモデルです。", + "url": "https://huggingface.co/recruit-jp/japanese-clip-vit-b-32-roberta-base", + "downloads": 37, + "description": "recruit-jp/japanese-clip-vit-b-32-roberta-baseOverviewDeveloped by: Recruit Co.", "source": "Hugging Face", - "score": 5.003023354807885e-06, - "project_name": "Mixtral-8x7B-Instruct-v0.1-japanese" + "score": 5.536009949257232e-06, + "project_name": "japanese-clip-vit-b-32-roberta-base" }, { - "url": "https://huggingface.co/watashiha/Watashiha-Llama-2-13B-Ogiri-sft", + "url": "https://huggingface.co/Aratako/ELYZA-japanese-Llama-2-MoE-2x7B-v0.1-GGUF", "downloads": 36, - "description": "The English document is here.", + "description": "ELYZA-japanese-Llama-2-MoE-2x7B-v0.1-GGUF概要Aratako/ELYZA-japanese-Llama-2-MoE-2x7B-v0.1の量子化済みGGUF版です。", "source": "Hugging Face", - "score": 5.003023354807885e-06, - "project_name": "Watashiha-Llama-2-13B-Ogiri-sft" + "score": 5.3863880587367665e-06, + "project_name": "ELYZA-japanese-Llama-2-MoE-2x7B-v0.1-GGUF" }, { - "url": "https://huggingface.co/rinna/japanese-stable-diffusion", + "url": "https://huggingface.co/wolf4032/bert-japanese-token-classification-search-local-cuisine", "downloads": 36, - "description": "One more step before getting this model.", + "description": "Model Card for Model ID料理を検索するための質問文から、検索検索用キーワードである固有表現を抽出しますModel DetailsModel Description例えば、「東京の肉料理で、春に食べられる、鶏肉を使った料理を教えてください」という文章を入力すると、「東京　→　都道府県/地方(AREA)」　「肉料理　→　種類(TYPE)」　「春　→　季節(SZN)」　「鶏肉　→　食材(INGR)」のように、固有表現を抽出します抽出対象は、AREA、TYPE、SZN、INGRの４つですLanguage(s) (NLP):", "source": "Hugging Face", - "score": 5.003023354807885e-06, - "project_name": "japanese-stable-diffusion" + "score": 5.3863880587367665e-06, + "project_name": "bert-japanese-token-classification-search-local-cuisine" }, { - "url": "https://huggingface.co/Local-Novel-LLM-project/Ninja-v1-GGUF", + "url": "https://huggingface.co/elyza/Llama-3-ELYZA-JP-8B", "downloads": 35, - "description": "Ninja-v1 のGGUF版Our Models for GGUFVecteus-GGUFNinja-v1-GGUFNinja-v1-NSFW-GGUFNinja-v1-128k-GGUFNinja-v1-NSFW-128k-GGUF", + "description": "Llama-3-ELYZA-JP-8BModel DescriptionLlama-3-ELYZA-JP-8B is a large language model trained by ELYZA, Inc.Based on meta-llama/Meta-Llama-3-8B-Instruct, it has been enhanced for Japanese usage through additional pre-training and instruction tuning.", "source": "Hugging Face", - "score": 4.864050483840999e-06, - "project_name": "Ninja-v1-GGUF" + "score": 5.2367661682163005e-06, + "project_name": "Llama-3-ELYZA-JP-8B" }, { - "url": "https://huggingface.co/HODACHI/mistral-seven-merged", + "url": "https://huggingface.co/ku-accms/roberta-base-japanese-ssuw", "downloads": 35, - "description": "概要「HODACHI/mistral-seven-merged」は、Mistral系の高い日本語性能を誇る、７つのモデルを再トレーニングを行いながら評価を行い、マージした結果に基づき生まれたモデルです。", + "description": "ku-accms/roberta-base-japanese-ssuwModel descriptionThis is a pre-trained Japanese RoBERTa base model for super short unit words (SSUW).", "source": "Hugging Face", - "score": 4.864050483840999e-06, - "project_name": "mistral-seven-merged" + "score": 5.2367661682163005e-06, + "project_name": "roberta-base-japanese-ssuw" }, { "url": "https://huggingface.co/ku-nlp/roberta-large-japanese-char-wwm", - "downloads": 35, + "downloads": 34, "description": "ku-nlp/roberta-large-japanese-char-wwmModel descriptionThis is a Japanese RoBERTa large model pre-trained on Japanese Wikipedia and the Japanese portion of CC-100.This model is trained with character-level tokenization and whole word masking.", "source": "Hugging Face", - "score": 4.864050483840999e-06, + "score": 5.0871442776958346e-06, "project_name": "roberta-large-japanese-char-wwm" }, { - "url": "https://huggingface.co/alabnii/jmedroberta-base-manbyo-wordpiece-vocab50000", - "downloads": 34, - "description": "alabnii/jmedroberta-base-manbyo-wordpiece-vocab50000Model descriptionThis is a Japanese RoBERTa base model pre-trained on academic articles in medical sciences collected by Japan Science and Technology Agency (JST).", - "source": "Hugging Face", - "score": 4.725077612874113e-06, - "project_name": "jmedroberta-base-manbyo-wordpiece-vocab50000" - }, - { - "url": "https://huggingface.co/llm-book/bert-base-japanese-v3-bpr-passage-aio", - "downloads": 34, - "description": "bert-base-japanese-v3-bpr-passage-aio「大規模言語モデル入門」の第9章で紹介している文書検索モデルBPRのパッセージエンコーダです。", + "url": "https://huggingface.co/vitouphy/wav2vec2-xls-r-300m-japanese", + "downloads": 33, + "description": "This model is for transcribing audio into Hiragana, one format of Japanese language.", "source": "Hugging Face", - "score": 4.725077612874113e-06, - "project_name": "bert-base-japanese-v3-bpr-passage-aio" + "score": 4.9375223871753694e-06, + "project_name": "wav2vec2-xls-r-300m-japanese" }, { - "url": "https://huggingface.co/retrieva-jp/t5-base-short", - "downloads": 34, - "description": "Model card for model IDThis is a T5 v1.1 model, pre-trained on a Japanese corpus.", + "url": "https://huggingface.co/mr4/bert-base-jp-sentiment-analysis", + "downloads": 33, + "description": "Sentiment Analysis in Japanese - Phân tích cảm xúc trong tiếng NhậtBert phân tích cảm xúcModel descriptionMô hình có tác dụng xác định cảm xúc của đoạn văn.", "source": "Hugging Face", - "score": 4.725077612874113e-06, - "project_name": "t5-base-short" + "score": 4.9375223871753694e-06, + "project_name": "bert-base-jp-sentiment-analysis" }, { - "url": "https://huggingface.co/webbigdata/C3TR-Adapter_gptq", + "url": "https://huggingface.co/aixsatoshi/Swallow-MX-8x7b-NVE-chatvector-Mixtral-instruct", "downloads": 33, - "description": "Model card英日、日英翻訳用モデルC3TR-AdapterのGPTQ4bit量子化版です。", + "description": "更新情報日本語機能とinstructベクトルのバランス調整したver.2をアップロードしましたSwallow-MX-8x7b-NVE-chatvector-Mixtral-instruct-v2モデル概要Swallow-MX-8x7b-NVE-v0.1に対し、Mixtral-8x7B-Instruct-v0.1とMixtral-8x7B-v0.1の差分をマージしたモデルです。", "source": "Hugging Face", - "score": 4.5861047419072275e-06, - "project_name": "C3TR-Adapter_gptq" + "score": 4.9375223871753694e-06, + "project_name": "Swallow-MX-8x7b-NVE-chatvector-Mixtral-instruct" }, { - "url": "https://huggingface.co/izumi-lab/bert-small-japanese-fin", - "downloads": 32, - "description": "BERT small Japanese financeThis is a BERT model pretrained on texts in the Japanese language.", + "url": "https://huggingface.co/kz/mt5base-finetuned-patentsum-japanese-small", + "downloads": 33, + "description": "Google's mt5-base fine-tuned in Japanese to summarize patent claims in a limited Pharmaceutical domain.", "source": "Hugging Face", - "score": 4.447131870940342e-06, - "project_name": "bert-small-japanese-fin" + "score": 4.9375223871753694e-06, + "project_name": "mt5base-finetuned-patentsum-japanese-small" }, { - "url": "https://huggingface.co/Formzu/bert-base-japanese-jsnli", + "url": "https://huggingface.co/rinna/japanese-data2vec-audio-base", "downloads": 32, - "description": "bert-base-japanese-jsnliThis model is a fine-tuned version of cl-tohoku/bert-base-japanese-v2 on the JSNLI dataset.", + "description": "rinna/japanese-data2vec-audio-baseOverviewThis is a Japanese data2vec Audio Base model trained by rinna Co.", "source": "Hugging Face", - "score": 4.447131870940342e-06, - "project_name": "bert-base-japanese-jsnli" + "score": 4.7879004966549035e-06, + "project_name": "japanese-data2vec-audio-base" }, { "url": "https://huggingface.co/line-corporation/japanese-large-lm-3.6b-instruction-sft-4bit-128g-actorder_False", "downloads": 32, "description": "japanese-large-lm-3.6b-instruction-sft-4bit-128g-actorder_FalseThis repository provides a 3.6B parameters Japanese language quantized model, fine-tuned and trained by LINE Corporation.", "source": "Hugging Face", - "score": 4.447131870940342e-06, + "score": 4.7879004966549035e-06, "project_name": "japanese-large-lm-3.6b-instruction-sft-4bit-128g-actorder_False" }, { - "url": "https://huggingface.co/DavidAU/alpaca-guanaco-japanese-gpt-1b-Q8_0-GGUF", + "url": "https://huggingface.co/SkelterLabsInc/bert-base-japanese-jaquad", "downloads": 32, - "description": "DavidAU/alpaca-guanaco-japanese-gpt-1b-Q8_0-GGUFThis model was converted to GGUF format from inu-ai/alpaca-guanaco-japanese-gpt-1b using llama.cpp via the ggml.ai's GGUF-my-repo space.", + "description": "BERT base Japanese - JaQuADDescriptionA Japanese Question Answering model fine-tuned on JaQuAD.Please refer BERT base Japanese for details about the pre-training model.", "source": "Hugging Face", - "score": 4.447131870940342e-06, - "project_name": "alpaca-guanaco-japanese-gpt-1b-Q8_0-GGUF" + "score": 4.7879004966549035e-06, + "project_name": "bert-base-japanese-jaquad" }, { - "url": "https://huggingface.co/schroneko/ELYZA-japanese-Llama-2-13b-fast-instruct-gguf", + "url": "https://huggingface.co/izumi-lab/bert-small-japanese-fin", "downloads": 31, - "description": "ELYZA-japanese-Llama-2-13b-fast-instruct-ggufELYZA-japanese-Llama-2-13b-fast-instructの GGUF", + "description": "BERT small Japanese financeThis is a BERT model pretrained on texts in the Japanese language.", "source": "Hugging Face", - "score": 4.3081589999734566e-06, - "project_name": "ELYZA-japanese-Llama-2-13b-fast-instruct-gguf" + "score": 4.6382786061344375e-06, + "project_name": "bert-small-japanese-fin" }, { - "url": "https://huggingface.co/uzabase/luke-japanese-wordpiece-base", - "downloads": 30, - "description": "studio-ousia/luke-japanese-baseに対して次の変更を加えたモデルです。", + "url": "https://huggingface.co/Mizuiro-sakura/luke-japanese-large-finetuned-QA", + "downloads": 31, + "description": "このモデルはluke-japanese-large-liteをファインチューニングして、Question-Answeringに用いれるようにしたものです。", "source": "Hugging Face", - "score": 4.16918612900657e-06, - "project_name": "luke-japanese-wordpiece-base" + "score": 4.6382786061344375e-06, + "project_name": "luke-japanese-large-finetuned-QA" }, { - "url": "https://huggingface.co/shinyice/chatvector-llava-v1.5-plus-houou-v3-7b", + "url": "https://huggingface.co/Spiral-AI/Spiral-RetNet-3b-base", + "downloads": 31, + "description": "SpiralAI Spiral-RetNet-3b-baseWe have conducted pre-training from scratch on the RetNet (https://arxiv.org/abs/2307.08621)", + "source": "Hugging Face", + "score": 4.6382786061344375e-06, + "project_name": "Spiral-RetNet-3b-base" + }, + { + "url": "https://huggingface.co/Aratako/ELYZA-japanese-Llama-2-fast-MoE-2x7B-v0.1-GGUF", "downloads": 30, - "description": "Chatvector-llava-v1.5-plus-Houou-v3-7b Model CardModel Details※好奇心から生まれたモデルです。", + "description": "ELYZA-japanese-Llama-2-fast-MoE-2x7B-v0.1-GGUF概要Aratako/ELYZA-japanese-Llama-2-fast-MoE-2x7B-v0.1の量子化済みGGUF版です。", "source": "Hugging Face", - "score": 4.16918612900657e-06, - "project_name": "chatvector-llava-v1.5-plus-houou-v3-7b" + "score": 4.4886567156139715e-06, + "project_name": "ELYZA-japanese-Llama-2-fast-MoE-2x7B-v0.1-GGUF" }, { - "url": "https://huggingface.co/kit-nlp/bert-base-japanese-basic-char-v2-cyberbullying", + "url": "https://huggingface.co/Local-Novel-LLM-project/Ninja-v1-GGUF", "downloads": 30, - "description": "electra-base-cyberbullyingThis is a BERT Base model for the Japanese language finetuned for automatic cyberbullying detection.", + "description": "Ninja-v1 のGGUF版Our Models for GGUFVecteus-GGUFNinja-v1-GGUFNinja-v1-NSFW-GGUFNinja-v1-128k-GGUFNinja-v1-NSFW-128k-GGUF", "source": "Hugging Face", - "score": 4.16918612900657e-06, - "project_name": "bert-base-japanese-basic-char-v2-cyberbullying" + "score": 4.4886567156139715e-06, + "project_name": "Ninja-v1-GGUF" }, { - "url": "https://huggingface.co/Formzu/roberta-base-japanese-jsnli", - "downloads": 29, - "description": "roberta-base-japanese-jsnliThis model is a fine-tuned version of nlp-waseda/roberta-base-japanese on the JSNLI dataset.", + "url": "https://huggingface.co/drewschaub/whisper-large-v3-japanese-4k-steps", + "downloads": 30, + "description": "whisper-large-v3-japanese-4k-stepsThis model is a fine-tuned version of openai/whisper-large-v3 on the Common Voice 16.1 dataset.", "source": "Hugging Face", - "score": 4.030213258039685e-06, - "project_name": "roberta-base-japanese-jsnli" + "score": 4.4886567156139715e-06, + "project_name": "whisper-large-v3-japanese-4k-steps" }, { - "url": "https://huggingface.co/Mizuiro-sakura/luke-japanese-large-finetuned-QA", - "downloads": 29, - "description": "このモデルはluke-japanese-large-liteをファインチューニングして、Question-Answeringに用いれるようにしたものです。", + "url": "https://huggingface.co/doc2query/msmarco-japanese-mt5-base-v1", + "downloads": 30, + "description": "doc2query/msmarco-japanese-mt5-base-v1This is a doc2query model based on mT5 (also known as docT5query).", "source": "Hugging Face", - "score": 4.030213258039685e-06, - "project_name": "luke-japanese-large-finetuned-QA" + "score": 4.4886567156139715e-06, + "project_name": "msmarco-japanese-mt5-base-v1" }, { - "url": "https://huggingface.co/Mizuiro-sakura/luke-japanese-base-lite-jsquad", + "url": "https://huggingface.co/oshizo/japanese-e5-mistral-7b_slerp", "downloads": 29, - "description": "このモデルはluke-japanese-base-liteをファインチューニングして、Question-Answeringに用いれるようにしたものです。", + "description": "This model was created by merging intfloat/e5-mistral-7b-instruct and stabilityai/japanese-stablelm-base-gamma-7b.", "source": "Hugging Face", - "score": 4.030213258039685e-06, - "project_name": "luke-japanese-base-lite-jsquad" + "score": 4.339034825093506e-06, + "project_name": "japanese-e5-mistral-7b_slerp" }, { - "url": "https://huggingface.co/jweb/japanese-soseki-gpt2-1b", - "downloads": 29, - "description": "japanese-soseki-gpt2-1bThis repository provides a 1.3B-parameter finetuned Japanese GPT2 model.", + "url": "https://huggingface.co/tokyotech-llm/Swallow-70b-NVE-instruct-hf", + "downloads": 28, + "description": "SwallowOur Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data.", "source": "Hugging Face", - "score": 4.030213258039685e-06, - "project_name": "japanese-soseki-gpt2-1b" + "score": 4.1894129345730404e-06, + "project_name": "Swallow-70b-NVE-instruct-hf" }, { "url": "https://huggingface.co/ku-nlp/gpt2-medium-japanese-char", "downloads": 28, "description": "Model Card for Japanese character-level", "source": "Hugging Face", - "score": 3.891240387072799e-06, + "score": 4.1894129345730404e-06, "project_name": "gpt2-medium-japanese-char" }, { - "url": "https://huggingface.co/nlp-waseda/gpt2-small-japanese", + "url": "https://huggingface.co/Helsinki-NLP/opus-mt-ja-it", "downloads": 27, - "description": "nlp-waseda/gpt2-small-japaneseThis model is Japanese GPT-2 pretrained on Japanese Wikipedia and CC-100.Intended uses & limitationsYou can use the raw model for text generation or fine-tune it to a downstream task.", + "description": "jpn-itasource group: Japanesetarget group: ItalianOPUS readme: jpn-itamodel: transformer-alignsource language(s): jpn jpn_Hani jpn_Hira jpn_Kana jpn_Latn jpn_Yiiitarget language(s): itamodel: transformer-alignpre-processing: normalization + SentencePiece (spm32k,spm32k)", "source": "Hugging Face", - "score": 3.7522675161059134e-06, - "project_name": "gpt2-small-japanese" + "score": 4.0397910440525745e-06, + "project_name": "opus-mt-ja-it" }, { - "url": "https://huggingface.co/NovelAI/genji-jp", + "url": "https://huggingface.co/KoichiYasuoka/roberta-base-japanese-aozora", "downloads": 27, - "description": "Genji-JP 6BPlease check our blog post for more details, samples, evaluations and more:BlogpostModel DescriptionGenji-JP 6B is a model finetuned on our Japanese storytelling dataset based on EleutherAI's GPT-J 6B model.", + "description": "roberta-base-japanese-aozoraModel DescriptionThis is a RoBERTa model pre-trained on 青空文庫 texts with Japanese-LUW-Tokenizer.", "source": "Hugging Face", - "score": 3.7522675161059134e-06, - "project_name": "genji-jp" + "score": 4.0397910440525745e-06, + "project_name": "roberta-base-japanese-aozora" }, { - "url": "https://huggingface.co/doc2query/msmarco-japanese-mt5-base-v1", + "url": "https://huggingface.co/ptaszynski/yacis-electra-small-japanese-cyberbullying", "downloads": 27, - "description": "doc2query/msmarco-japanese-mt5-base-v1This is a doc2query model based on mT5 (also known as docT5query).", + "description": "yacis-electra-small-cyberbullyingThis is an ELECTRA Small model for the Japanese language finetuned for automatic cyberbullying detection.", "source": "Hugging Face", - "score": 3.7522675161059134e-06, - "project_name": "msmarco-japanese-mt5-base-v1" + "score": 4.0397910440525745e-06, + "project_name": "yacis-electra-small-japanese-cyberbullying" }, { - "url": "https://huggingface.co/Local-Novel-LLM-project/Vecteus-V2-7B", + "url": "https://huggingface.co/alabnii/jmedroberta-base-manbyo-wordpiece-vocab50000", "downloads": 27, - "description": "Vecteus-V2-7Bこのモデルは、ベクトルマージなどを用い作成された高性能ベースモデルです。 ", + "description": "alabnii/jmedroberta-base-manbyo-wordpiece-vocab50000Model descriptionThis is a Japanese RoBERTa base model pre-trained on academic articles in medical sciences collected by Japan Science and Technology Agency (JST).", "source": "Hugging Face", - "score": 3.7522675161059134e-06, - "project_name": "Vecteus-V2-7B" + "score": 4.0397910440525745e-06, + "project_name": "jmedroberta-base-manbyo-wordpiece-vocab50000" }, { - "url": "https://huggingface.co/Spiral-AI/Spiral-RetNet-3b-base", + "url": "https://huggingface.co/Formzu/roberta-base-japanese-jsnli", "downloads": 26, - "description": "SpiralAI Spiral-RetNet-3b-baseWe have conducted pre-training from scratch on the RetNet (https://arxiv.org/abs/2307.08621)", + "description": "roberta-base-japanese-jsnliThis model is a fine-tuned version of nlp-waseda/roberta-base-japanese on the JSNLI dataset.", "source": "Hugging Face", - "score": 3.613294645139028e-06, - "project_name": "Spiral-RetNet-3b-base" + "score": 3.890169153532109e-06, + "project_name": "roberta-base-japanese-jsnli" }, { - "url": "https://huggingface.co/Momerio/meigen_generate_Japanese", + "url": "https://huggingface.co/watashiha/Watashiha-Llama-2-13B-Ogiri-sft", "downloads": 26, - "description": "名言推論モデル", + "description": "The English document is here.", "source": "Hugging Face", - "score": 3.613294645139028e-06, - "project_name": "meigen_generate_Japanese" + "score": 3.890169153532109e-06, + "project_name": "Watashiha-Llama-2-13B-Ogiri-sft" }, { - "url": "https://huggingface.co/Tomohiro/RealMedNLP_CR_JA", + "url": "https://huggingface.co/retrieva-jp/t5-base-short", "downloads": 25, - "description": "This is a model for named entity recognition of Japanese medical documents.", + "description": "Model card for model IDThis is a T5 v1.1 model, pre-trained on a Japanese corpus.", "source": "Hugging Face", - "score": 3.474321774172142e-06, - "project_name": "RealMedNLP_CR_JA" + "score": 3.7405472630116434e-06, + "project_name": "t5-base-short" }, { - "url": "https://huggingface.co/huranokuma/es", + "url": "https://huggingface.co/nlp-waseda/roberta_jtruthfulqa", "downloads": 25, - "description": "ESを書くAIJapanese GPT-2 modelをファインチューニングしましたファインチューニングには、内定者の二万件以上のESを用いました。", - "source": "Hugging Face", - "score": 3.474321774172142e-06, - "project_name": "es" - }, - { - "url": "https://huggingface.co/Mizuiro-sakura/t5-CAMERA-title-generation", - "downloads": 24, - "description": "sonoisa/t5-base-japaneseをファインチューニングして、タイトル生成に用いれるようにしたモデルです。", + "description": "Finetuned Waseda RoBERTa to evaluate the generated answers on JTruthfulQA.", "source": "Hugging Face", - "score": 3.3353489032052565e-06, - "project_name": "t5-CAMERA-title-generation" + "score": 3.7405472630116434e-06, + "project_name": "roberta_jtruthfulqa" }, { - "url": "https://huggingface.co/knok/japanese-distilgpt2", - "downloads": 23, - "description": "日本語 gpt2 蒸留モデルこのモデルはrinna/japanese-gpt2-meduimを教師として蒸留したものです。", + "url": "https://huggingface.co/uzabase/luke-japanese-wordpiece-base", + "downloads": 25, + "description": "studio-ousia/luke-japanese-baseに対して次の変更を加えたモデルです。", "source": "Hugging Face", - "score": 3.1963760322383706e-06, - "project_name": "japanese-distilgpt2" + "score": 3.7405472630116434e-06, + "project_name": "luke-japanese-wordpiece-base" }, { - "url": "https://huggingface.co/alabnii/jmedroberta-base-sentencepiece", + "url": "https://huggingface.co/schroneko/ELYZA-japanese-Llama-2-13b-fast-instruct-gguf", "downloads": 23, - "description": "alabnii/jmedroberta-base-sentencepieceModel descriptionThis is a Japanese RoBERTa base model pre-trained on academic articles in medical sciences collected by Japan Science and Technology Agency (JST).", + "description": "ELYZA-japanese-Llama-2-13b-fast-instruct-ggufELYZA-japanese-Llama-2-13b-fast-instructの GGUF", "source": "Hugging Face", - "score": 3.1963760322383706e-06, - "project_name": "jmedroberta-base-sentencepiece" + "score": 3.441303481970712e-06, + "project_name": "ELYZA-japanese-Llama-2-13b-fast-instruct-gguf" }, { - "url": "https://huggingface.co/nu-dialogue/sfc2022-stable-diffusion", + "url": "https://huggingface.co/fukugawa/transformer-lm-japanese-0.1b", "downloads": 23, - "description": "SFCOCO Stable Diffusion Model CardSFCOCO Stable Diffusion is a Japanese-specific latent text-to-image diffusion model capable of generating photo-realistic images given any text input.", - "source": "Hugging Face", - "score": 3.1963760322383706e-06, - "project_name": "sfc2022-stable-diffusion" - }, - { - "url": "https://huggingface.co/KoichiYasuoka/bert-large-japanese-upos", - "downloads": 22, - "description": "bert-large-japanese-uposModel DescriptionThis is a BERT model pre-trained on Japanese Wikipedia texts for POS-tagging and dependency-parsing, derived from bert-large-japanese-char-extended.", + "description": "transformer-lm-japanese-0.1bThis is a JAX/Flax-based transformer language model trained on a Japanese dataset.", "source": "Hugging Face", - "score": 3.057403161271485e-06, - "project_name": "bert-large-japanese-upos" + "score": 3.441303481970712e-06, + "project_name": "transformer-lm-japanese-0.1b" }, { - "url": "https://huggingface.co/snu-nia-12/wav2vec2-xls-r-300m_nia12_phone-hiragana_japanese", + "url": "https://huggingface.co/alabnii/jmedroberta-base-sentencepiece", "downloads": 22, - "description": "Wav2Vec2-XLS-R-300M-Japanese-HiraganaFine-tuned facebook/wav2vec2-xls-r-300m on Japanese Hiragana characters using JSUT, JVS, Common Voice, and in-house dataset.", + "description": "alabnii/jmedroberta-base-sentencepieceModel descriptionThis is a Japanese RoBERTa base model pre-trained on academic articles in medical sciences collected by Japan Science and Technology Agency (JST).", "source": "Hugging Face", - "score": 3.057403161271485e-06, - "project_name": "wav2vec2-xls-r-300m_nia12_phone-hiragana_japanese" + "score": 3.2916815914502463e-06, + "project_name": "jmedroberta-base-sentencepiece" }, { - "url": "https://huggingface.co/akiFQC/bert-base-japanese-v3_nli-jsnli-jnli-jsick", + "url": "https://huggingface.co/alabnii/jmedroberta-base-manbyo-wordpiece", "downloads": 22, - "description": "Cross-Encoder for Natural Language Inference(NLI) for JapaneseThis model was trained using SentenceTransformers Cross-Encoder class.", + "description": "alabnii/jmedroberta-base-manbyo-wordpieceModel descriptionThis is a Japanese RoBERTa base model pre-trained on academic articles in medical sciences collected by Japan Science and Technology Agency (JST).", "source": "Hugging Face", - "score": 3.057403161271485e-06, - "project_name": "bert-base-japanese-v3_nli-jsnli-jnli-jsick" + "score": 3.2916815914502463e-06, + "project_name": "jmedroberta-base-manbyo-wordpiece" }, { - "url": "https://huggingface.co/Local-Novel-LLM-project/Ocuteus-v1", + "url": "https://huggingface.co/KoichiYasuoka/deberta-small-japanese-upos", "downloads": 22, - "description": "VecteusをベースにLLavaに対応させたモデルです。", + "description": "deberta-small-japanese-uposModel DescriptionThis is a DeBERTa(V2) model pre-trained on 青空文庫 texts for POS-tagging and dependency-parsing, derived from deberta-small-japanese-aozora.", "source": "Hugging Face", - "score": 3.057403161271485e-06, - "project_name": "Ocuteus-v1" + "score": 3.2916815914502463e-06, + "project_name": "deberta-small-japanese-upos" }, { - "url": "https://huggingface.co/drewschaub/whisper-large-v3-japanese-4k-steps", + "url": "https://huggingface.co/Momerio/meigen_generate_Japanese", "downloads": 22, - "description": "whisper-large-v3-japanese-4k-stepsThis model is a fine-tuned version of openai/whisper-large-v3 on the Common Voice 16.1 dataset.", + "description": "名言推論モデル", "source": "Hugging Face", - "score": 3.057403161271485e-06, - "project_name": "whisper-large-v3-japanese-4k-steps" + "score": 3.2916815914502463e-06, + "project_name": "meigen_generate_Japanese" }, { - "url": "https://huggingface.co/mr4/bert-base-jp-sentiment-analysis", + "url": "https://huggingface.co/abeja/Mixtral-8x7B-Instruct-v0.1-japanese", "downloads": 22, - "description": "Sentiment Analysis in Japanese - Phân tích cảm xúc trong tiếng NhậtBert phân tích cảm xúcModel descriptionMô hình có tác dụng xác định cảm xúc của đoạn văn.", - "source": "Hugging Face", - "score": 3.057403161271485e-06, - "project_name": "bert-base-jp-sentiment-analysis" - }, - { - "url": "https://huggingface.co/alabnii/jmedroberta-base-manbyo-wordpiece", - "downloads": 21, - "description": "alabnii/jmedroberta-base-manbyo-wordpieceModel descriptionThis is a Japanese RoBERTa base model pre-trained on academic articles in medical sciences collected by Japan Science and Technology Agency (JST).", + "description": "Mixtral-8x7B-Instruct-v0.1-japaneseMixtral-8x7B-Instruct-v0.1-japaneseはMixtral-8x7B-Instruct-v0.1をベースに日本語の語彙拡張継続事前学���を実施したモデルです。", "source": "Hugging Face", - "score": 2.9184302903045992e-06, - "project_name": "jmedroberta-base-manbyo-wordpiece" + "score": 3.2916815914502463e-06, + "project_name": "Mixtral-8x7B-Instruct-v0.1-japanese" }, { - "url": "https://huggingface.co/offtoung/tsukuyomi-chan-calm2-7b", + "url": "https://huggingface.co/nlp-waseda/roberta-large-japanese-with-auto-jumanpp", "downloads": 21, - "description": "つくよみちゃんデータセットを用いて calm-2-7b-chat をファインチューニングしたモデルです。", + "description": "nlp-waseda/roberta-large-japanese-with-auto-jumanppModel descriptionThis is a Japanese RoBERTa large model pretrained on Japanese Wikipedia and the Japanese portion of CC-100.How to useYou can use this model for masked language modeling as follows:from transformers import AutoTokenizer, AutoModelForMaskedLMtokenizer = AutoTokenizer.from_pretrained(\"nlp-waseda/roberta-large-japanese-with-auto-jumanpp\")", "source": "Hugging Face", - "score": 2.9184302903045992e-06, - "project_name": "tsukuyomi-chan-calm2-7b" + "score": 3.1420597009297803e-06, + "project_name": "roberta-large-japanese-with-auto-jumanpp" }, { - "url": "https://huggingface.co/tokyotech-llm/Swallow-70b-NVE-hf", + "url": "https://huggingface.co/snu-nia-12/wav2vec2-xls-r-300m_nia12_phone-hiragana_japanese", "downloads": 21, - "description": "SwallowOur Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data.", + "description": "Wav2Vec2-XLS-R-300M-Japanese-HiraganaFine-tuned facebook/wav2vec2-xls-r-300m on Japanese Hiragana characters using JSUT, JVS, Common Voice, and in-house dataset.", "source": "Hugging Face", - "score": 2.9184302903045992e-06, - "project_name": "Swallow-70b-NVE-hf" + "score": 3.1420597009297803e-06, + "project_name": "wav2vec2-xls-r-300m_nia12_phone-hiragana_japanese" }, { - "url": "https://huggingface.co/sonoisa/t5-base-japanese-mC4-Wikipedia", + "url": "https://huggingface.co/espnet/kan-bayashi_jsut_vits_prosody", "downloads": 21, - "description": "日本語T5事前学習済みモデルThis is a T5 (Text-to-Text Transfer Transformer) model pretrained on Japanese corpus.", - "source": "Hugging Face", - "score": 2.9184302903045992e-06, - "project_name": "t5-base-japanese-mC4-Wikipedia" - }, - { - "url": "https://huggingface.co/izumi-lab/electra-base-japanese-discriminator", - "downloads": 20, - "description": "ELECTRA base Japanese discriminatorThis is a ELECTRA model pretrained on texts in the Japanese language.", - "source": "Hugging Face", - "score": 2.7794574193377138e-06, - "project_name": "electra-base-japanese-discriminator" - }, - { - "url": "https://huggingface.co/hiroshi-matsuda-rit/bert-base-japanese-basic-char-v2", - "downloads": 20, - "description": "BERT base Japanese (character-level tokenization with whole word masking, jawiki-20200831)This pretrained model is almost the same as cl-tohoku/bert-base-japanese-char-v2 but do not need fugashi or unidic_lite.", + "description": "ESPnet2 TTS pretrained modelkan-bayashi/jsut_vits_prosody♻", "source": "Hugging Face", - "score": 2.7794574193377138e-06, - "project_name": "bert-base-japanese-basic-char-v2" + "score": 3.1420597009297803e-06, + "project_name": "kan-bayashi_jsut_vits_prosody" }, { - "url": "https://huggingface.co/stockmark/bart-base-japanese-news", + "url": "https://huggingface.co/Mizuiro-sakura/luke-japanese-base-finetuned-jnli", "downloads": 20, - "description": "bart-base-japanese-news(base-sized model)This repository provides a Japanese BART model.", + "description": "このモデルはluke-japanese-baseをファインチューニングして、JNLI(文章の関係性判別)に用いれるようにしたものです。", "source": "Hugging Face", - "score": 2.7794574193377138e-06, - "project_name": "bart-base-japanese-news" + "score": 2.9924378104093148e-06, + "project_name": "luke-japanese-base-finetuned-jnli" }, { - "url": "https://huggingface.co/yellowback/gpt-neo-japanese-1.3B", - "downloads": 20, - "description": "GPT-Neo 1.3B pre-trained model for JapaneseModel DescriptionGPT2/GPT3 like model trained on Japanese.corpus.", + "url": "https://huggingface.co/neoai-inc/Llama-3-neoAI-8B-Chat-v0.1", + "downloads": 19, + "description": "Llama 3 neoAI 8B Chat v0.1Model DescriptionLlama 3 neoAI 8B Chat v0.1は，Meta-Llama-3-8B-Instructをベースとして日本語能力を強化するために事後学習を行なったモデルです．", "source": "Hugging Face", - "score": 2.7794574193377138e-06, - "project_name": "gpt-neo-japanese-1.3B" + "score": 2.842815919888849e-06, + "project_name": "Llama-3-neoAI-8B-Chat-v0.1" }, { - "url": "https://huggingface.co/astremo/friendly_JA", - "downloads": 20, - "description": "friendly_JA-Model　(T5 fine-tuned model)MT model trained using the friendly_JA Corpus attempting to make Japanese easier/more accessible to occidental people by using the Latin/English derived katakana lexicon instead of the standard Sino-Japanese lexiconExamplesinputoutput最適化を応用した機械翻訳モデルは高精度だオプティマイゼーションを応用したマシントランスレーションモデルは高いアキュラシーだ彼は架空の世界に住んでいる彼はイマジナリー世界に住んでいる新型コロナウイルスに感染してしまったコロナウイルスにかかってしまった深層学習は難しいディープラーニングはむずかしい新たな概念を紹介する新しいコンセプトを紹介する津波の警報が流れたツナミのアラートが流れた南海トラフの災害は震源地による南海トラフのディザスターはエピセンターによる息子は際どい内容の本を", + "url": "https://huggingface.co/Kendamarron/fineweb-edu-classifier-ja", + "downloads": 19, + "description": "HuggingFaceFW/fineweb-edu-classifierを再現するために、日本語データでpkshatech/GLuCoSE-base-jaを学習したモデルです。", "source": "Hugging Face", - "score": 2.7794574193377138e-06, - "project_name": "friendly_JA" + "score": 2.842815919888849e-06, + "project_name": "fineweb-edu-classifier-ja" }, { - "url": "https://huggingface.co/megagonlabs/transformers-ud-japanese-electra-base-discriminator", + "url": "https://huggingface.co/tohoku-nlp/bert-large-japanese-char", "downloads": 19, - "description": "transformers-ud-japanese-electra-ginza (sudachitra-wordpiece, mC4 Japanese) -", + "description": "BERT large Japanese (character-level tokenization with whole word masking, jawiki-20200831)This is a BERT model pretrained on texts in the Japanese language.", "source": "Hugging Face", - "score": 2.640484548370828e-06, - "project_name": "transformers-ud-japanese-electra-base-discriminator" + "score": 2.842815919888849e-06, + "project_name": "bert-large-japanese-char" }, { - "url": "https://huggingface.co/nlp-waseda/roberta-large-japanese-seq512-with-auto-jumanpp", + "url": "https://huggingface.co/offtoung/tsukuyomi-chan-calm2-7b", "downloads": 19, - "description": "nlp-waseda/roberta-large-japanese-seq512-with-auto-jumanppModel descriptionThis is a Japanese RoBERTa large model pretrained on Japanese Wikipedia and the Japanese portion of CC-100 with the maximum sequence length of 512.How to useYou can use this model for masked language modeling as follows:from transformers import AutoTokenizer, AutoModelForMaskedLMtokenizer = AutoTokenizer.from_pretrained(\"nlp-waseda/roberta-large-japanese-seq512-with-auto-jumanpp\")", + "description": "つくよみちゃんデータセットを用いて calm-2-7b-chat をファインチューニングしたモデルです。", "source": "Hugging Face", - "score": 2.640484548370828e-06, - "project_name": "roberta-large-japanese-seq512-with-auto-jumanpp" + "score": 2.842815919888849e-06, + "project_name": "tsukuyomi-chan-calm2-7b" }, { - "url": "https://huggingface.co/bardsai/finance-sentiment-ja-base", + "url": "https://huggingface.co/nlp-waseda/gpt2-small-japanese", "downloads": 19, - "description": "Finance Sentiment JA (base)Finance Sentiment JA (base) is a model based on bert-base-japanese for analyzing sentiment of Japanese financial news.", + "description": "nlp-waseda/gpt2-small-japaneseThis model is Japanese GPT-2 pretrained on Japanese Wikipedia and CC-100.Intended uses & limitationsYou can use the raw model for text generation or fine-tune it to a downstream task.", "source": "Hugging Face", - "score": 2.640484548370828e-06, - "project_name": "finance-sentiment-ja-base" + "score": 2.842815919888849e-06, + "project_name": "gpt2-small-japanese" }, { - "url": "https://huggingface.co/bennexx/cl-tohoku-bert-base-japanese-v3-jlpt-classifier", + "url": "https://huggingface.co/megagonlabs/t5-base-japanese-web-8k", "downloads": 19, - "description": "SummaryThis is a text classifier for assigning a JLPT level.", + "description": "t5-base-japanese-web-8k (with Byte-fallback, 8K)Descriptionmegagonlabs/t5-base-japanese-web-8k is a T5 (Text-to-Text Transfer Transformer) model pre-trained on Japanese web texts.", "source": "Hugging Face", - "score": 2.640484548370828e-06, - "project_name": "cl-tohoku-bert-base-japanese-v3-jlpt-classifier" + "score": 2.842815919888849e-06, + "project_name": "t5-base-japanese-web-8k" }, { - "url": "https://huggingface.co/espnet/kan-bayashi_jsut_vits_prosody", + "url": "https://huggingface.co/abhishek/autonlp-japanese-sentiment-59362", "downloads": 19, - "description": "ESPnet2 TTS pretrained modelkan-bayashi/jsut_vits_prosody♻", + "description": "Model Trained Using AutoNLPProblem type: Binary ClassificationModel ID: 59362Validation MetricsLoss: 0.13092292845249176Accuracy: 0.9527127414314258Precision: 0.9634070704982427Recall: 0.9842171959602166AUC: 0.9667289746092403F1:", "source": "Hugging Face", - "score": 2.640484548370828e-06, - "project_name": "kan-bayashi_jsut_vits_prosody" + "score": 2.842815919888849e-06, + "project_name": "autonlp-japanese-sentiment-59362" }, { - "url": "https://huggingface.co/Kendamarron/fineweb-edu-classifier-ja", + "url": "https://huggingface.co/ttop324/wav2vec2-live-japanese", "downloads": 18, - "description": "HuggingFaceFW/fineweb-edu-classifierを再現するために、日本語データでpkshatech/GLuCoSE-base-jaを学習したモデルです。", + "description": "wav2vec2-live-japanesehttps://github.com/ttop32/wav2vec2-live-japanese-translatorFine-tuned facebook/wav2vec2-large-xlsr-53 on Japanese hiragana using thecommon_voiceJSUTCSS10TEDxJP-10KJVSJSSSInference#usageimport torchimport torchaudiofrom datasets import load_datasetfrom transformers import Wav2Vec2ForCTC, Wav2Vec2Processormodel = Wav2Vec2ForCTC.from_pretrained(\"ttop324/wav2vec2-live-japanese\")", "source": "Hugging Face", - "score": 2.5015116774039424e-06, - "project_name": "fineweb-edu-classifier-ja" + "score": 2.6931940293683833e-06, + "project_name": "wav2vec2-live-japanese" }, { - "url": "https://huggingface.co/Helsinki-NLP/opus-mt-ja-it", + "url": "https://huggingface.co/stockmark/bart-base-japanese-news", "downloads": 18, - "description": "jpn-itasource group: Japanesetarget group: ItalianOPUS readme: jpn-itamodel: transformer-alignsource language(s): jpn jpn_Hani jpn_Hira jpn_Kana jpn_Latn jpn_Yiiitarget language(s): itamodel: transformer-alignpre-processing: normalization + SentencePiece (spm32k,spm32k)", + "description": "bart-base-japanese-news(base-sized model)This repository provides a Japanese BART model.", "source": "Hugging Face", - "score": 2.5015116774039424e-06, - "project_name": "opus-mt-ja-it" + "score": 2.6931940293683833e-06, + "project_name": "bart-base-japanese-news" }, { - "url": "https://huggingface.co/KoichiYasuoka/bert-large-japanese-unidic-luw-upos", + "url": "https://huggingface.co/k-ush/xlm-roberta-base-ance-en-jp-warmup", "downloads": 18, - "description": "bert-large-japanese-unidic-luw-uposModel DescriptionThis is a BERT model pre-trained on Japanese Wikipedia texts for POS-tagging and dependency-parsing, derived from bert-large-japanese.", + "description": "k-ush/xlm-roberta-base-ance-en-jp-warmupA XLM-RoBERTa-base model trained on mMARCO Japanese dataset with ANCE warmup script.", "source": "Hugging Face", - "score": 2.5015116774039424e-06, - "project_name": "bert-large-japanese-unidic-luw-upos" + "score": 2.6931940293683833e-06, + "project_name": "xlm-roberta-base-ance-en-jp-warmup" }, { - "url": "https://huggingface.co/skytnt/gpt2-japanese-lyric-medium", + "url": "https://huggingface.co/DavidAU/alpaca-guanaco-japanese-gpt-1b-Q8_0-GGUF", "downloads": 18, - "description": "Japanese GPT2 Lyric ModelModel descriptionThe model is used to generate Japanese lyrics.", + "description": "DavidAU/alpaca-guanaco-japanese-gpt-1b-Q8_0-GGUFThis model was converted to GGUF format from inu-ai/alpaca-guanaco-japanese-gpt-1b using llama.cpp via the ggml.ai's GGUF-my-repo space.", "source": "Hugging Face", - "score": 2.5015116774039424e-06, - "project_name": "gpt2-japanese-lyric-medium" + "score": 2.6931940293683833e-06, + "project_name": "alpaca-guanaco-japanese-gpt-1b-Q8_0-GGUF" }, { - "url": "https://huggingface.co/tohoku-nlp/stable-diffusion-xl-jp-base-1.0", + "url": "https://huggingface.co/tohoku-nlp/stable-diffusion-xl-jp-refiner-1.0", "downloads": 18, "description": "(English part follows Japanese one.", "source": "Hugging Face", - "score": 2.5015116774039424e-06, - "project_name": "stable-diffusion-xl-jp-base-1.0" + "score": 2.6931940293683833e-06, + "project_name": "stable-diffusion-xl-jp-refiner-1.0" }, { - "url": "https://huggingface.co/KoichiYasuoka/roberta-base-japanese-aozora-ud-head", + "url": "https://huggingface.co/ce-lery/dolly-japanese-gpt-1b-clone", "downloads": 18, - "description": "roberta-base-japanese-aozora-ud-headModel", + "description": "dolly-japanese-gpt-1b-clone概要rinna社の「japanese-gpt-1b」を、日本語データセット「databricks-dolly-15k-ja」を使用して学習させた推論モデルです。", "source": "Hugging Face", - "score": 2.5015116774039424e-06, - "project_name": "roberta-base-japanese-aozora-ud-head" + "score": 2.6931940293683833e-06, + "project_name": "dolly-japanese-gpt-1b-clone" }, { - "url": "https://huggingface.co/Bagus/wav2vec2-xlsr-japanese-speech-emotion-recognition", + "url": "https://huggingface.co/KoichiYasuoka/deberta-large-japanese-wikipedia-ud-goeswith", "downloads": 18, - "description": "This is for (private) DEMO only.", + "description": "deberta-large-japanese-wikipedia-ud-goeswithModel DescriptionThis is a DeBERTa(V2) model pretrained on Japanese Wikipedia and 青空文庫 texts for POS-tagging and dependency-parsing (using goeswith for subwords), derived from deberta-large-japanese-wikipedia-luw-upos and UD_Japanese-GSDLUW.How to Useclass UDgoeswith(object):def __init__(self,bert):", + "source": "Hugging Face", + "score": 2.6931940293683833e-06, + "project_name": "deberta-large-japanese-wikipedia-ud-goeswith" + }, + { + "url": "https://huggingface.co/KoichiYasuoka/bert-large-japanese-upos", + "downloads": 17, + "description": "bert-large-japanese-uposModel DescriptionThis is a BERT model pre-trained on Japanese Wikipedia texts for POS-tagging and dependency-parsing, derived from bert-large-japanese-char-extended.", "source": "Hugging Face", - "score": 2.5015116774039424e-06, - "project_name": "wav2vec2-xlsr-japanese-speech-emotion-recognition" + "score": 2.5435721388479173e-06, + "project_name": "bert-large-japanese-upos" }, { "url": "https://huggingface.co/nlp-waseda/gpt2-small-japanese-wikipedia", "downloads": 17, "description": "nlp-waseda/gpt2-small-japanese-wikipediaThis model is Japanese GPT-2 pretrained on Japanese Wikipedia.", "source": "Hugging Face", - "score": 2.3625388064370565e-06, + "score": 2.5435721388479173e-06, "project_name": "gpt2-small-japanese-wikipedia" }, { - "url": "https://huggingface.co/microsoft/unihanlm-base", + "url": "https://huggingface.co/megagonlabs/roberta-long-japanese", "downloads": 17, - "description": "Unihan LM: Coarse-to-Fine Chinese-Japanese Language Model Pretraining with the Unihan DatabaseModel descriptionChinese and Japanese share many characters with similar surface morphology.", + "description": "roberta-long-japanese (jumanpp + sentencepiece, mC4 Japanese)This is the longer input version of RoBERTa Japanese model pretrained on approximately 200", "source": "Hugging Face", - "score": 2.3625388064370565e-06, - "project_name": "unihanlm-base" + "score": 2.5435721388479173e-06, + "project_name": "roberta-long-japanese" }, { - "url": "https://huggingface.co/TylorShine/distilhubert-ft-japanese-50k", + "url": "https://huggingface.co/webbigdata/C3TR-Adapter_gptq", "downloads": 17, - "description": "distilhubert-ft-japanese-50kFine-tuned (more precisely, continue trained)", + "description": "Model card英日、日英翻訳用モデルC3TR-AdapterのGPTQ4bit量子化版です。", "source": "Hugging Face", - "score": 2.3625388064370565e-06, - "project_name": "distilhubert-ft-japanese-50k" + "score": 2.5435721388479173e-06, + "project_name": "C3TR-Adapter_gptq" }, { - "url": "https://huggingface.co/KoichiYasuoka/roberta-large-japanese-aozora-ud-head", + "url": "https://huggingface.co/jweb/japanese-soseki-gpt2-1b", "downloads": 17, - "description": "roberta-large-japanese-aozora-ud-headModel", + "description": "japanese-soseki-gpt2-1bThis repository provides a 1.3B-parameter finetuned Japanese GPT2 model.", "source": "Hugging Face", - "score": 2.3625388064370565e-06, - "project_name": "roberta-large-japanese-aozora-ud-head" + "score": 2.5435721388479173e-06, + "project_name": "japanese-soseki-gpt2-1b" }, { - "url": "https://huggingface.co/espnet/kan-bayashi_jvs_jvs010_vits_prosody", + "url": "https://huggingface.co/hiroshi-matsuda-rit/bert-base-japanese-basic-char-v2", "downloads": 16, - "description": "ESPnet2 TTS pretrained modelkan-bayashi/jvs_jvs010_vits_prosody♻", + "description": "BERT base Japanese (character-level tokenization with whole word masking, jawiki-20200831)This pretrained model is almost the same as cl-tohoku/bert-base-japanese-char-v2 but do not need fugashi or unidic_lite.", "source": "Hugging Face", - "score": 2.223565935470171e-06, - "project_name": "kan-bayashi_jvs_jvs010_vits_prosody" + "score": 2.3939502483274517e-06, + "project_name": "bert-base-japanese-basic-char-v2" }, { - "url": "https://huggingface.co/ptaszynski/yacis-electra-small-japanese-cyberbullying", + "url": "https://huggingface.co/AbeShinzo0708/Japanese-Starling-ChatV-7B-exl2", "downloads": 16, - "description": "yacis-electra-small-cyberbullyingThis is an ELECTRA Small model for the Japanese language finetuned for automatic cyberbullying detection.", + "description": "Japanese-Starling-ChatV-7Bこのモデルは\"chatntq-ja-7b-v1.0\"をベースにした7Bパラメータの日本語チャットモデルです。", "source": "Hugging Face", - "score": 2.223565935470171e-06, - "project_name": "yacis-electra-small-japanese-cyberbullying" + "score": 2.3939502483274517e-06, + "project_name": "Japanese-Starling-ChatV-7B-exl2" }, { - "url": "https://huggingface.co/Mizuiro-sakura/luke-japanese-base-finetuned-jnli", + "url": "https://huggingface.co/Akimite/Qwen2-7b-Instruct-Boku-v3", "downloads": 16, - "description": "このモデルはluke-japanese-baseをファインチューニングして、JNLI(文章の関係性判別)に用いれるようにしたものです。", + "description": "Akimite/Qwen2-7b-Instruct-Boku-v2のマイナーチェンジ版です。", "source": "Hugging Face", - "score": 2.223565935470171e-06, - "project_name": "luke-japanese-base-finetuned-jnli" + "score": 2.3939502483274517e-06, + "project_name": "Qwen2-7b-Instruct-Boku-v3" }, { - "url": "https://huggingface.co/minkhantycc/translation-en-ja", + "url": "https://huggingface.co/Aratako/Oumuamua-7b-instruct-v2-RP", "downloads": 16, - "description": "This model is the fine-tuned version of Helsinki-NLP/opus-mt-ja-en on bsd_ja_en dataset.", + "description": "Oumuamua-7b-instruct-v2-RPnitky/Oumuamua-7b-instruct-v2をロールプレイ用にLoRAでファインチューニングしたモデルです。", "source": "Hugging Face", - "score": 2.223565935470171e-06, - "project_name": "translation-en-ja" + "score": 2.3939502483274517e-06, + "project_name": "Oumuamua-7b-instruct-v2-RP" }, { - "url": "https://huggingface.co/eepj/wstcg-mt-ja-en", + "url": "https://huggingface.co/alfredplpl/sarashina2-7b-it", "downloads": 16, - "description": "WS TCG Card Text TranslatorA Japanese-English machine translation model specifically trained for translating card text from the Weiss Schwarz (WS) Trading Card Game, fine-tuned on Helsinki-NLP/opus-mt-ja-en.", + "description": "Sarashina2-7B Instructsarashina2-7Bを会話できるようにフルファインチューニングしたものです。", "source": "Hugging Face", - "score": 2.223565935470171e-06, - "project_name": "wstcg-mt-ja-en" + "score": 2.3939502483274517e-06, + "project_name": "sarashina2-7b-it" }, { - "url": "https://huggingface.co/Akimite/Qwen2-7b-Instruct-Boku-v3", + "url": "https://huggingface.co/paulhindemith/fasttext-jp-embedding", "downloads": 16, - "description": "Akimite/Qwen2-7b-Instruct-Boku-v2のマイナーチェンジ版です。", + "description": "fasttext-jp-embeddingThis model is experimental.", "source": "Hugging Face", - "score": 2.223565935470171e-06, - "project_name": "Qwen2-7b-Instruct-Boku-v3" + "score": 2.3939502483274517e-06, + "project_name": "fasttext-jp-embedding" }, { - "url": "https://huggingface.co/k-ush/xlm-roberta-base-ance-en-jp-warmup", + "url": "https://huggingface.co/skytnt/gpt2-japanese-lyric-medium", "downloads": 16, - "description": "k-ush/xlm-roberta-base-ance-en-jp-warmupA XLM-RoBERTa-base model trained on mMARCO Japanese dataset with ANCE warmup script.", + "description": "Japanese GPT2 Lyric ModelModel descriptionThe model is used to generate Japanese lyrics.", "source": "Hugging Face", - "score": 2.223565935470171e-06, - "project_name": "xlm-roberta-base-ance-en-jp-warmup" + "score": 2.3939502483274517e-06, + "project_name": "gpt2-japanese-lyric-medium" }, { - "url": "https://huggingface.co/astremo/JAINU", + "url": "https://huggingface.co/NovelAI/genji-jp", "downloads": 16, - "description": "JAINU-Model　(T5 fine-tuned model)JAINU is a Japanese - Ainu language machine translation model.", + "description": "Genji-JP 6BPlease check our blog post for more details, samples, evaluations and more:BlogpostModel DescriptionGenji-JP 6B is a model finetuned on our Japanese storytelling dataset based on EleutherAI's GPT-J 6B model.", "source": "Hugging Face", - "score": 2.223565935470171e-06, - "project_name": "JAINU" + "score": 2.3939502483274517e-06, + "project_name": "genji-jp" }, { "url": "https://huggingface.co/izumi-lab/electra-small-japanese-discriminator", "downloads": 15, "description": "ELECTRA small Japanese discriminatorThis is a ELECTRA model pretrained on texts in the Japanese language.", "source": "Hugging Face", - "score": 2.084593064503285e-06, + "score": 2.2443283578069858e-06, "project_name": "electra-small-japanese-discriminator" }, { - "url": "https://huggingface.co/naclbit/gpt-j-japanese-6.8b", - "downloads": 15, - "description": "This pre-trained model is work in progress!", - "source": "Hugging Face", - "score": 2.084593064503285e-06, - "project_name": "gpt-j-japanese-6.8b" - }, - { - "url": "https://huggingface.co/ttop324/wav2vec2-live-japanese", + "url": "https://huggingface.co/wietsedv/xlm-roberta-base-ft-udpos28-ja", "downloads": 15, - "description": "wav2vec2-live-japanesehttps://github.com/ttop32/wav2vec2-live-japanese-translatorFine-tuned facebook/wav2vec2-large-xlsr-53 on Japanese hiragana using thecommon_voiceJSUTCSS10TEDxJP-10KJVSJSSSInference#usageimport torchimport torchaudiofrom datasets import load_datasetfrom transformers import Wav2Vec2ForCTC, Wav2Vec2Processormodel = Wav2Vec2ForCTC.from_pretrained(\"ttop324/wav2vec2-live-japanese\")", + "description": "XLM-RoBERTa base Universal Dependencies v2.8 POS tagging:", "source": "Hugging Face", - "score": 2.084593064503285e-06, - "project_name": "wav2vec2-live-japanese" + "score": 2.2443283578069858e-06, + "project_name": "xlm-roberta-base-ft-udpos28-ja" }, { "url": "https://huggingface.co/kit-nlp/bert-base-japanese-basic-char-v2-irony", "downloads": 15, "description": "bert-base-ironyThis is a BERT Base model for the Japanese language finetuned for automatic irony detection.", "source": "Hugging Face", - "score": 2.084593064503285e-06, + "score": 2.2443283578069858e-06, "project_name": "bert-base-japanese-basic-char-v2-irony" }, { - "url": "https://huggingface.co/omzn/facemark_detection", - "downloads": 15, - "description": "Facemark DetectionThis model classifies given text into facemark(1) or not(0).", - "source": "Hugging Face", - "score": 2.084593064503285e-06, - "project_name": "facemark_detection" - }, - { - "url": "https://huggingface.co/alfredplpl/sarashina2-7b-it", - "downloads": 15, - "description": "Sarashina2-7B Instructsarashina2-7Bを会話できるようにフルファインチューニングしたものです。", - "source": "Hugging Face", - "score": 2.084593064503285e-06, - "project_name": "sarashina2-7b-it" - }, - { - "url": "https://huggingface.co/oshizo/japanese-sexual-moderation-v2", - "downloads": 15, - "description": "japanese-sexual-moderation-v2は、studio-ousia/luke-japanese-large-liteをファインチューニングしたモデルです。", - "source": "Hugging Face", - "score": 2.084593064503285e-06, - "project_name": "japanese-sexual-moderation-v2" - }, - { - "url": "https://huggingface.co/KoichiYasuoka/deberta-base-japanese-wikipedia-ud-goeswith", + "url": "https://huggingface.co/inu-ai/alpaca-guanaco-japanese-gpt-1b", "downloads": 15, - "description": "deberta-base-japanese-wikipedia-ud-goeswithModel DescriptionThis is a DeBERTa(V2) model pretrained on Japanese Wikipedia and 青空文庫 texts for POS-tagging and dependency-parsing (using goeswith for subwords), derived from deberta-base-japanese-wikipedia-luw-upos and UD_Japanese-GSDLUW.How to Useclass UDgoeswith(object):def __init__(self,bert):", + "description": "alpaca-guanaco-japanese-gpt-1b1.3Bパラメータの日本語GPTモデルを使用した対話AIです。", "source": "Hugging Face", - "score": 2.084593064503285e-06, - "project_name": "deberta-base-japanese-wikipedia-ud-goeswith" + "score": 2.2443283578069858e-06, + "project_name": "alpaca-guanaco-japanese-gpt-1b" }, { - "url": "https://huggingface.co/megagonlabs/roberta-long-japanese", + "url": "https://huggingface.co/Jumtra/mpt-7b-base", "downloads": 15, - "description": "roberta-long-japanese (jumanpp + sentencepiece, mC4 Japanese)This is the longer input version of RoBERTa Japanese model pretrained on approximately 200", + "description": "MPT-7B-baseこのモデルは、MosaicMLのllm-foundryリポジトリを使用してmosaicml/mpt-7bをファインチューニングしたモデルです。", "source": "Hugging Face", - "score": 2.084593064503285e-06, - "project_name": "roberta-long-japanese" + "score": 2.2443283578069858e-06, + "project_name": "mpt-7b-base" }, { - "url": "https://huggingface.co/wietsedv/xlm-roberta-base-ft-udpos28-ja", + "url": "https://huggingface.co/nu-dialogue/sfc2022-stable-diffusion", "downloads": 15, - "description": "XLM-RoBERTa base Universal Dependencies v2.8 POS tagging:", + "description": "SFCOCO Stable Diffusion Model CardSFCOCO Stable Diffusion is a Japanese-specific latent text-to-image diffusion model capable of generating photo-realistic images given any text input.", "source": "Hugging Face", - "score": 2.084593064503285e-06, - "project_name": "xlm-roberta-base-ft-udpos28-ja" + "score": 2.2443283578069858e-06, + "project_name": "sfc2022-stable-diffusion" }, { - "url": "https://huggingface.co/ClassCat/gpt2-base-japanese-v2", + "url": "https://huggingface.co/naclbit/gpt-j-japanese-6.8b", "downloads": 14, - "description": "GPT2 Japanese base model version 2Prerequisitestransformers==4.19.2Model", + "description": "This pre-trained model is work in progress!", "source": "Hugging Face", - "score": 1.9456201935363996e-06, - "project_name": "gpt2-base-japanese-v2" + "score": 2.0947064672865202e-06, + "project_name": "gpt-j-japanese-6.8b" }, { - "url": "https://huggingface.co/Mizuiro-sakura/luke-japanese-base-commonsenseqa", + "url": "https://huggingface.co/kit-nlp/bert-base-japanese-basic-char-v2-cyberbullying", "downloads": 14, - "description": "このモデルはluke-japanese-baseをファインチューニングして、JCommonsenseQA(選択式応答)に用いれるようにしたものです。", + "description": "electra-base-cyberbullyingThis is a BERT Base model for the Japanese language finetuned for automatic cyberbullying detection.", "source": "Hugging Face", - "score": 1.9456201935363996e-06, - "project_name": "luke-japanese-base-commonsenseqa" + "score": 2.0947064672865202e-06, + "project_name": "bert-base-japanese-basic-char-v2-cyberbullying" }, { - "url": "https://huggingface.co/ku-accms/bert-base-japanese-ssuw", + "url": "https://huggingface.co/nlp-waseda/roberta-large-japanese-seq512-with-auto-jumanpp", "downloads": 14, - "description": "ku-accms/bert-base-japanese-ssuwModel descriptionThis is a pre-trained Japanese BERT base model for super short unit words (SSUW).", + "description": "nlp-waseda/roberta-large-japanese-seq512-with-auto-jumanppModel descriptionThis is a Japanese RoBERTa large model pretrained on Japanese Wikipedia and the Japanese portion of CC-100 with the maximum sequence length of 512.How to useYou can use this model for masked language modeling as follows:from transformers import AutoTokenizer, AutoModelForMaskedLMtokenizer = AutoTokenizer.from_pretrained(\"nlp-waseda/roberta-large-japanese-seq512-with-auto-jumanpp\")", "source": "Hugging Face", - "score": 1.9456201935363996e-06, - "project_name": "bert-base-japanese-ssuw" + "score": 2.0947064672865202e-06, + "project_name": "roberta-large-japanese-seq512-with-auto-jumanpp" }, { - "url": "https://huggingface.co/Jumtra/mpt-7b-inst", + "url": "https://huggingface.co/retrieva-jp/t5-large-medium", "downloads": 14, - "description": "MPT-7B-instこのモデルは、MosaicMLのllm-foundryリポジトリを使用してmosaicml/mpt-7b-instructをファインチューニングしたモデルです。", + "description": "Model card for model IDThis is a T5 v1.1 model, pre-trained on a Japanese corpus.", "source": "Hugging Face", - "score": 1.9456201935363996e-06, - "project_name": "mpt-7b-inst" + "score": 2.0947064672865202e-06, + "project_name": "t5-large-medium" }, { - "url": "https://huggingface.co/hitachi-nlp/bert-base-japanese_nothing-wordpiece", + "url": "https://huggingface.co/hitachi-nlp/bert-base-japanese_jumanpp-bpe", "downloads": 14, - "description": "Japanese BERT-base (Nothing + WordPiece)How to load the tokenizerPlease download the dictionary file for Nothing + WordPiece from our GitHub repository.", + "description": "Japanese BERT-base (Juman++ + BPE)How to load the tokenizerPlease download the dictionary file for Juman++ + BPE from our GitHub repository.", "source": "Hugging Face", - "score": 1.9456201935363996e-06, - "project_name": "bert-base-japanese_nothing-wordpiece" + "score": 2.0947064672865202e-06, + "project_name": "bert-base-japanese_jumanpp-bpe" }, { - "url": "https://huggingface.co/haih2/open-calm-7b-summarizer-lora", + "url": "https://huggingface.co/abeja/Mixtral-8x7B-Instruct-v0.1-japanese-alpha-merged", "downloads": 14, - "description": "Fine-tuned OpenCALM-7B Adapters for Meeting SummarizationDescriptionThese are weights for LoRA adapters fine-tuned on the OpenCALM-7B (Andonian et al.", + "description": "Mixtral-8x7B-Instruct-v0.1-japanese-alpha-mergedMixtral-8x7B-Instruct-v0.1-japanese-alpha-mergedはMixtral-8x7B-Instruct-v0.1をベースに日本語の語彙拡張継続事前学習を実施した学習途中のモデルに対して、差分マージを実施したモデルです。", "source": "Hugging Face", - "score": 1.9456201935363996e-06, - "project_name": "open-calm-7b-summarizer-lora" + "score": 2.0947064672865202e-06, + "project_name": "Mixtral-8x7B-Instruct-v0.1-japanese-alpha-merged" }, { - "url": "https://huggingface.co/sambanovasystems/SambaLingo-Japanese-Base", + "url": "https://huggingface.co/abeja/Mixtral-8x7B-v0.1-japanese", "downloads": 14, - "description": "SambaLingo-Japanese-BaseSambaLingo-Japanese-Base is a pretrained Bi-lingual Japanese and English model that adapts Llama-2-7b to Japanese by training on 42 billion tokens from the Japanese split of the Cultura-X dataset.", + "description": "Mixtral-8x7B-v0.1-japaneseMixtral-8x7B-v0.1-japaneseはMixtral-8x7B-v0.1をベースに日本語の語彙拡張継続事前学習を実施したモデルです。", "source": "Hugging Face", - "score": 1.9456201935363996e-06, - "project_name": "SambaLingo-Japanese-Base" + "score": 2.0947064672865202e-06, + "project_name": "Mixtral-8x7B-v0.1-japanese" }, { - "url": "https://huggingface.co/tokyotech-llm/Swallow-70b-NVE-instruct-hf", + "url": "https://huggingface.co/bardsai/finance-sentiment-ja-base", "downloads": 14, - "description": "SwallowOur Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data.", + "description": "Finance Sentiment JA (base)Finance Sentiment JA (base) is a model based on bert-base-japanese for analyzing sentiment of Japanese financial news.", "source": "Hugging Face", - "score": 1.9456201935363996e-06, - "project_name": "Swallow-70b-NVE-instruct-hf" + "score": 2.0947064672865202e-06, + "project_name": "finance-sentiment-ja-base" }, { - "url": "https://huggingface.co/Mizuiro-sakura/deberta-v2-large-japanese-finetuned-ner", + "url": "https://huggingface.co/hitachi-nlp/bert-base-japanese_sudachi-unigram", "downloads": 14, - "description": "このモデルはdeberta-v2-large-japaneseをファインチューニングして固有表現抽出（NER）に用いれるようにしたものです。", + "description": "Japanese BERT-base (Sudachi + Unigram)How to load the tokenizerPlease download the dictionary file for Sudachi + Unigram from our GitHub repository.", "source": "Hugging Face", - "score": 1.9456201935363996e-06, - "project_name": "deberta-v2-large-japanese-finetuned-ner" + "score": 2.0947064672865202e-06, + "project_name": "bert-base-japanese_sudachi-unigram" }, { - "url": "https://huggingface.co/megagonlabs/t5-base-japanese-web-8k", + "url": "https://huggingface.co/hitachi-nlp/bert-base-japanese_sudachi-bpe", "downloads": 14, - "description": "t5-base-japanese-web-8k (with Byte-fallback, 8K)Descriptionmegagonlabs/t5-base-japanese-web-8k is a T5 (Text-to-Text Transfer Transformer) model pre-trained on Japanese web texts.", + "description": "Japanese BERT-base (Sudachi + BPE)How to load the tokenizerPlease download the dictionary file for Sudachi + BPE from our GitHub repository.", "source": "Hugging Face", - "score": 1.9456201935363996e-06, - "project_name": "t5-base-japanese-web-8k" + "score": 2.0947064672865202e-06, + "project_name": "bert-base-japanese_sudachi-bpe" }, { - "url": "https://huggingface.co/Helsinki-NLP/opus-mt-ja-nl", + "url": "https://huggingface.co/Jumtra/mpt-7b-inst", "downloads": 14, - "description": "jpn-nldsource group: Japanesetarget group: DutchOPUS readme: jpn-nldmodel: transformer-alignsource language(s): jpn jpn_Hani jpn_Hira jpn_Kana jpn_Latntarget language(s): nldmodel: transformer-alignpre-processing: normalization + SentencePiece (spm32k,spm32k)", - "source": "Hugging Face", - "score": 1.9456201935363996e-06, - "project_name": "opus-mt-ja-nl" - }, - { - "url": "https://huggingface.co/KoichiYasuoka/roberta-base-japanese-char-luw-upos", - "downloads": 13, - "description": "roberta-base-japanese-char-luw-uposModel", - "source": "Hugging Face", - "score": 1.806647322569514e-06, - "project_name": "roberta-base-japanese-char-luw-upos" - }, - { - "url": "https://huggingface.co/KoichiYasuoka/roberta-base-japanese-aozora-char", - "downloads": 13, - "description": "roberta-base-japanese-aozora-charModel DescriptionThis is a RoBERTa model pre-trained on 青空文庫 texts with character tokenizer.", - "source": "Hugging Face", - "score": 1.806647322569514e-06, - "project_name": "roberta-base-japanese-aozora-char" - }, - { - "url": "https://huggingface.co/tohoku-nlp/bert-large-japanese-char", - "downloads": 13, - "description": "BERT large Japanese (character-level tokenization with whole word masking, jawiki-20200831)This is a BERT model pretrained on texts in the Japanese language.", - "source": "Hugging Face", - "score": 1.806647322569514e-06, - "project_name": "bert-large-japanese-char" - }, - { - "url": "https://huggingface.co/ptaszynski/yacis-electra-small-japanese", - "downloads": 13, - "description": "yacis-electra-smallThis is ELECTRA Small model for Japanese pretrained on 354 million sentences / 5.6 billion words of YACIS blog corpus.", - "source": "Hugging Face", - "score": 1.806647322569514e-06, - "project_name": "yacis-electra-small-japanese" - }, - { - "url": "https://huggingface.co/kit-nlp/transformers-ud-japanese-electra-base-discriminator-irony", - "downloads": 13, - "description": "Electra Base Japanese IronyThis is an ELECTRA Base model for the Japanese language finetuned for automatic irony detection.", - "source": "Hugging Face", - "score": 1.806647322569514e-06, - "project_name": "transformers-ud-japanese-electra-base-discriminator-irony" - }, - { - "url": "https://huggingface.co/Jumtra/mpt-7b-base", - "downloads": 13, - "description": "MPT-7B-baseこのモデルは、MosaicMLのllm-foundryリポジトリを使用してmosaicml/mpt-7bをファインチューニングしたモデルです。", - "source": "Hugging Face", - "score": 1.806647322569514e-06, - "project_name": "mpt-7b-base" - }, - { - "url": "https://huggingface.co/hitachi-nlp/bert-base-japanese_jumanpp-unigram", - "downloads": 13, - "description": "Japanese BERT-base (Juman++ + Unigram)How to load the tokenizerPlease download the dictionary file for Juman++ + Unigram from our GitHub repository.", - "source": "Hugging Face", - "score": 1.806647322569514e-06, - "project_name": "bert-base-japanese_jumanpp-unigram" - }, - { - "url": "https://huggingface.co/Deepreneur/blue-lizard", - "downloads": 13, - "description": "Deepreneur-blue-lizardModel DescriptionDeepreneur-blue-lizardは、MetaのLlama-2-7bに対して、Wikipediaや書籍等の日本語の学習データを用いて追加事前学習と独自データによるファインチューニングを実施したモデルです。", - "source": "Hugging Face", - "score": 1.806647322569514e-06, - "project_name": "blue-lizard" - }, - { - "url": "https://huggingface.co/MuneK/bert-large-japanese-v2-finetuned-jed", - "downloads": 13, - "description": "bert-large-japanese-v2-finetuned-wrimeThis model is finetuned from cl-tohoku/bert-large-japanese-v2 by JEmpatheticDialogues.", - "source": "Hugging Face", - "score": 1.806647322569514e-06, - "project_name": "bert-large-japanese-v2-finetuned-jed" - }, - { - "url": "https://huggingface.co/ebisuke/liz-nojaloli-nxja-ja", - "downloads": 13, - "description": "ebisuke/liz-nojaloli-nxja-jaLicenseMITベースとしてabeja/gpt-neox-japanese-2.7bを使用しています。", - "source": "Hugging Face", - "score": 1.806647322569514e-06, - "project_name": "liz-nojaloli-nxja-ja" - }, - { - "url": "https://huggingface.co/hyperonym/barba", - "downloads": 13, - "description": "BarbaBarba is a multilingual natural language inference model for textual entailment and zero-shot text classification, available as an end-to-end service through TensorFlow Serving.", - "source": "Hugging Face", - "score": 1.806647322569514e-06, - "project_name": "barba" - }, - { - "url": "https://huggingface.co/KoichiYasuoka/roberta-large-japanese-luw-upos", - "downloads": 12, - "description": "roberta-large-japanese-luw-uposModel", + "description": "MPT-7B-instこのモデルは、MosaicMLのllm-foundryリポジトリを使用してmosaicml/mpt-7b-instructをファインチューニングしたモデルです。", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "roberta-large-japanese-luw-upos" + "score": 2.0947064672865202e-06, + "project_name": "mpt-7b-inst" }, { - "url": "https://huggingface.co/KoichiYasuoka/roberta-large-japanese-char-luw-upos", - "downloads": 12, - "description": "roberta-large-japanese-char-luw-uposModel", + "url": "https://huggingface.co/TylorShine/distilhubert-ft-japanese-50k", + "downloads": 14, + "description": "distilhubert-ft-japanese-50kFine-tuned (more precisely, continue trained)", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "roberta-large-japanese-char-luw-upos" + "score": 2.0947064672865202e-06, + "project_name": "distilhubert-ft-japanese-50k" }, { - "url": "https://huggingface.co/KoichiYasuoka/roberta-base-japanese-luw-upos", - "downloads": 12, - "description": "roberta-base-japanese-luw-uposModel", + "url": "https://huggingface.co/Mizuiro-sakura/luke-japanese-base-lite-jsquad", + "downloads": 14, + "description": "このモデルはluke-japanese-base-liteをファインチューニングして、Question-Answeringに用いれるようにしたものです。", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "roberta-base-japanese-luw-upos" + "score": 2.0947064672865202e-06, + "project_name": "luke-japanese-base-lite-jsquad" }, { - "url": "https://huggingface.co/huranokuma/es2", - "downloads": 12, - "description": "ESを書くAIJapanese GPT-2", + "url": "https://huggingface.co/omzn/facemark_detection", + "downloads": 14, + "description": "Facemark DetectionThis model classifies given text into facemark(1) or not(0).", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "es2" + "score": 2.0947064672865202e-06, + "project_name": "facemark_detection" }, { - "url": "https://huggingface.co/KoichiYasuoka/deberta-large-japanese-aozora-ud-goeswith", - "downloads": 12, - "description": "deberta-large-japanese-aozora-ud-goeswithModel DescriptionThis is a DeBERTa(V2) model pretrained on 青空文庫 texts for POS-tagging and dependency-parsing (using goeswith for subwords), derived from deberta-large-japanese-luw-upos and UD_Japanese-GSDLUW.How to Useclass UDgoeswith(object):def __init__(self,bert):", + "url": "https://huggingface.co/ptaszynski/yacis-electra-small-japanese", + "downloads": 13, + "description": "yacis-electra-smallThis is ELECTRA Small model for Japanese pretrained on 354 million sentences / 5.6 billion words of YACIS blog corpus.", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "deberta-large-japanese-aozora-ud-goeswith" + "score": 1.9450845767660547e-06, + "project_name": "yacis-electra-small-japanese" }, { - "url": "https://huggingface.co/KoichiYasuoka/deberta-base-japanese-aozora-ud-goeswith", - "downloads": 12, - "description": "deberta-base-japanese-aozora-ud-goeswithModel DescriptionThis is a DeBERTa(V2) model pretrained on 青空文庫 texts for POS-tagging and dependency-parsing (using goeswith for subwords), derived from deberta-base-japanese-aozora and UD_Japanese-GSDLUW.How to Useclass UDgoeswith(object):def __init__(self,bert):", + "url": "https://huggingface.co/kit-nlp/transformers-ud-japanese-electra-base-discriminator-irony", + "downloads": 13, + "description": "Electra Base Japanese IronyThis is an ELECTRA Base model for the Japanese language finetuned for automatic irony detection.", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "deberta-base-japanese-aozora-ud-goeswith" + "score": 1.9450845767660547e-06, + "project_name": "transformers-ud-japanese-electra-base-discriminator-irony" }, { - "url": "https://huggingface.co/Mizuiro-sakura/bert-large-japanese-v2-finetuned-ner", - "downloads": 12, - "description": "このモデルはcl-tohoku/bert-large-japanese-v2をファインチューニングして、固有表現抽出（NER）に用いれるようにしたものです。", + "url": "https://huggingface.co/Mizuiro-sakura/deberta-v2-base-juman-finetuned-commonsenseqa", + "downloads": 13, + "description": "このモデルはdeberta-v2-base-japaneseをファインチューニングしてCommonsenseQA(選択式の質問)に用いれるようにしたものです。", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "bert-large-japanese-v2-finetuned-ner" + "score": 1.9450845767660547e-06, + "project_name": "deberta-v2-base-juman-finetuned-commonsenseqa" }, { - "url": "https://huggingface.co/hitachi-nlp/bert-base-japanese_nothing-bpe", - "downloads": 12, - "description": "Japanese BERT-base (Nothing + BPE)How to load the tokenizerPlease download the dictionary file for Nothing + BPE from our GitHub repository.", + "url": "https://huggingface.co/Mizuiro-sakura/luke-japanese-base-commonsenseqa", + "downloads": 13, + "description": "このモデルはluke-japanese-baseをファインチューニングして、JCommonsenseQA(選択式応答)に用いれるようにしたものです。", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "bert-base-japanese_nothing-bpe" + "score": 1.9450845767660547e-06, + "project_name": "luke-japanese-base-commonsenseqa" }, { - "url": "https://huggingface.co/hitachi-nlp/bert-base-japanese_vaporetto-wordpiece", - "downloads": 12, - "description": "Japanese BERT-base (Vaporetto + WordPiece)How to load the tokenizerPlease download the dictionary file for Vaporetto + WordPiece from our GitHub repository.", + "url": "https://huggingface.co/ebisuke/liz-nojaloli-nxja-ja", + "downloads": 13, + "description": "ebisuke/liz-nojaloli-nxja-jaLicenseMITベースとしてabeja/gpt-neox-japanese-2.7bを使用しています。", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "bert-base-japanese_vaporetto-wordpiece" + "score": 1.9450845767660547e-06, + "project_name": "liz-nojaloli-nxja-ja" }, { - "url": "https://huggingface.co/hitachi-nlp/bert-base-japanese_sudachi-unigram", - "downloads": 12, - "description": "Japanese BERT-base (Sudachi + Unigram)How to load the tokenizerPlease download the dictionary file for Sudachi + Unigram from our GitHub repository.", + "url": "https://huggingface.co/mathewthe2/manga-ocr-base", + "downloads": 13, + "description": "Manga OCROptical character recognition for Japanese text, with the main focus being Japanese manga.", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "bert-base-japanese_sudachi-unigram" + "score": 1.9450845767660547e-06, + "project_name": "manga-ocr-base" }, { - "url": "https://huggingface.co/hitachi-nlp/bert-base-japanese_jumanpp-wordpiece", - "downloads": 12, - "description": "Japanese BERT-base (Juman++ + WordPiece)How to load the tokenizerPlease download the dictionary file for Juman++ +", + "url": "https://huggingface.co/Kendamarron/Tokara-0.5B-Chat-v0.1", + "downloads": 13, + "description": "モデルについてQwen/Qwen1.5-0.5Bを日英データ5Bトークンで継続事前学習したTokara-0.5B-v0.1にchat vectorで対話能力を加えたモデルになります。", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "bert-base-japanese_jumanpp-wordpiece" + "score": 1.9450845767660547e-06, + "project_name": "Tokara-0.5B-Chat-v0.1" }, { - "url": "https://huggingface.co/tohoku-nlp/stable-diffusion-xl-jp-refiner-1.0", - "downloads": 12, - "description": "(English part follows Japanese one.", + "url": "https://huggingface.co/haih2/open-calm-7b-summarizer-lora", + "downloads": 13, + "description": "Fine-tuned OpenCALM-7B Adapters for Meeting SummarizationDescriptionThese are weights for LoRA adapters fine-tuned on the OpenCALM-7B (Andonian et al.", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "stable-diffusion-xl-jp-refiner-1.0" + "score": 1.9450845767660547e-06, + "project_name": "open-calm-7b-summarizer-lora" }, { - "url": "https://huggingface.co/tarudesu/gendec-with-distilmbert", - "downloads": 12, - "description": "INPUT: Japanese name in ROMAJI FORMOUTPUT:", + "url": "https://huggingface.co/hitachi-nlp/bert-base-japanese_nothing-wordpiece", + "downloads": 13, + "description": "Japanese BERT-base (Nothing + WordPiece)How to load the tokenizerPlease download the dictionary file for Nothing + WordPiece from our GitHub repository.", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "gendec-with-distilmbert" + "score": 1.9450845767660547e-06, + "project_name": "bert-base-japanese_nothing-wordpiece" }, { - "url": "https://huggingface.co/sarulab-speech/hubert-base-jtube", - "downloads": 12, - "description": "hubert-base-jtubeThis repo provides model weights for the hubert-base model trained on the JTubeSpeech corpus.Scroll down for the model usageFAQQ. ", + "url": "https://huggingface.co/hyperonym/barba", + "downloads": 13, + "description": "BarbaBarba is a multilingual natural language inference model for textual entailment and zero-shot text classification, available as an end-to-end service through TensorFlow Serving.", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "hubert-base-jtube" + "score": 1.9450845767660547e-06, + "project_name": "barba" }, { - "url": "https://huggingface.co/KoichiYasuoka/deberta-v3-base-japanese-ud-goeswith", - "downloads": 12, - "description": "deberta-v3-base-japanese-ud-goeswithModel DescriptionThis is a DeBERTa(V3) model pretrained on LLM-jp corpus v1.0 for POS-tagging and dependency-parsing (using goeswith for subwords), derived from deberta-v3-base-japanese and UD_Japanese-GSDLUW.How to Useclass UDgoeswith(object):def __init__(self,bert):", + "url": "https://huggingface.co/ku-accms/bert-base-japanese-ssuw", + "downloads": 13, + "description": "ku-accms/bert-base-japanese-ssuwModel descriptionThis is a pre-trained Japanese BERT base model for super short unit words (SSUW).", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "deberta-v3-base-japanese-ud-goeswith" + "score": 1.9450845767660547e-06, + "project_name": "bert-base-japanese-ssuw" }, { - "url": "https://huggingface.co/abeja/Mixtral-8x7B-Instruct-v0.1-japanese-alpha-merged", - "downloads": 12, - "description": "Mixtral-8x7B-Instruct-v0.1-japanese-alpha-mergedMixtral-8x7B-Instruct-v0.1-japanese-alpha-mergedはMixtral-8x7B-Instruct-v0.1をベースに日本語の語彙拡張継続事前学習を実施した学習途中のモデルに対して、差分マージを実施したモデルです。", + "url": "https://huggingface.co/KoichiYasuoka/roberta-large-japanese-juman-ud-goeswith", + "downloads": 13, + "description": "roberta-large-japanese-juman-ud-goeswithModel DescriptionThis is a RoBERTa model pretrained on Japanese Wikipedia and CC-100 texts for POS-tagging and dependency-parsing (using goeswith for subwords), derived from roberta-large-japanese.", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "Mixtral-8x7B-Instruct-v0.1-japanese-alpha-merged" + "score": 1.9450845767660547e-06, + "project_name": "roberta-large-japanese-juman-ud-goeswith" }, { - "url": "https://huggingface.co/A-Funakoshi/bert-base-japanese-v3-wrime-v1", - "downloads": 12, - "description": "ベースモデル：cl-tohoku/bert-base-japanese-whole-word-maskingデータセット：llm-book/wrime-sentimentオプティマイザ: adafactorOptunaでハイパーパラメータ探索学習率スケジュールのタイプ(lr_scheduler_type):", + "url": "https://huggingface.co/KoichiYasuoka/bert-base-japanese-char-extended", + "downloads": 13, + "description": "bert-base-japanese-char-extendedModel", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "bert-base-japanese-v3-wrime-v1" + "score": 1.9450845767660547e-06, + "project_name": "bert-base-japanese-char-extended" }, { - "url": "https://huggingface.co/c299m/japanese_stock_sentiment", - "downloads": 12, - "description": "Japanese Stock Comment Sentiment ModelThis model is a sentiment analysis tool specifically trained to analyze comments and discussions related to Japanese stocks.", + "url": "https://huggingface.co/spow12/Visual-novel-transcriptor", + "downloads": 13, + "description": "Model Card for Model IDFine tunned ASR model from distil-whisper/distil-large-v2.This model aimed to transcribe japanese audio especially visual novel.", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "japanese_stock_sentiment" + "score": 1.9450845767660547e-06, + "project_name": "Visual-novel-transcriptor" }, { - "url": "https://huggingface.co/arc-r/faster-whisper-large-v2-mix-jp", + "url": "https://huggingface.co/izumi-lab/electra-small-paper-japanese-discriminator", "downloads": 12, - "description": "whisper-large-v2-mix-jp model for CTranslate2This repository contains the conversion of vumichien/whisper-large-v2-mix-jp to the CTranslate2 model format.", + "description": "ELECTRA small Japanese discriminatorThis is a ELECTRA model pretrained on texts in the Japanese language.", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "faster-whisper-large-v2-mix-jp" + "score": 1.7954626862455887e-06, + "project_name": "electra-small-paper-japanese-discriminator" }, { - "url": "https://huggingface.co/hitachi-nlp/bert-base-japanese_jumanpp-bpe", + "url": "https://huggingface.co/Mizuiro-sakura/luke-large-commonsenseqa-japanese", "downloads": 12, - "description": "Japanese BERT-base (Juman++ + BPE)How to load the tokenizerPlease download the dictionary file for Juman++ + BPE from our GitHub repository.", + "description": "このモデルはluke-japanese-largeをファインチューニングして、JCommonsenseQA(選択式応答)に用いれるようにしたものです。", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "bert-base-japanese_jumanpp-bpe" + "score": 1.7954626862455887e-06, + "project_name": "luke-large-commonsenseqa-japanese" }, { "url": "https://huggingface.co/Mizuiro-sakura/luke-japanese-large-finetuned-ner", "downloads": 12, "description": "このモデルはluke-japanese-largeをファインチューニングして、固有表現抽出（NER）に用いれるようにしたものです。", "source": "Hugging Face", - "score": 1.6676744516026283e-06, + "score": 1.7954626862455887e-06, "project_name": "luke-japanese-large-finetuned-ner" }, { - "url": "https://huggingface.co/retrieva-jp/t5-large-medium", + "url": "https://huggingface.co/Mizuiro-sakura/deberta-v2-large-japanese-finetuned-ner", "downloads": 12, - "description": "Model card for model IDThis is a T5 v1.1 model, pre-trained on a Japanese corpus.", + "description": "このモデルはdeberta-v2-large-japaneseをファインチューニングして固有表現抽出（NER）に用いれるようにしたものです。", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "t5-large-medium" + "score": 1.7954626862455887e-06, + "project_name": "deberta-v2-large-japanese-finetuned-ner" }, { - "url": "https://huggingface.co/Mizuiro-sakura/luke-large-commonsenseqa-japanese", + "url": "https://huggingface.co/c299m/japanese_stock_sentiment", "downloads": 12, - "description": "このモデルはluke-japanese-largeをファインチューニングして、JCommonsenseQA(選択式応答)に用いれるようにしたものです。", + "description": "Japanese Stock Comment Sentiment ModelThis model is a sentiment analysis tool specifically trained to analyze comments and discussions related to Japanese stocks.", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "luke-large-commonsenseqa-japanese" + "score": 1.7954626862455887e-06, + "project_name": "japanese_stock_sentiment" }, { - "url": "https://huggingface.co/Mizuiro-sakura/deberta-v2-base-japanese-finetuned-ner", + "url": "https://huggingface.co/oshizo/japanese-sexual-moderation-v2", "downloads": 12, - "description": "このモデルはdeberta-v2-base-japaneseをファインチューニングして固有表現抽出（NER）に用いれるようにしたものです。", + "description": "japanese-sexual-moderation-v2は、studio-ousia/luke-japanese-large-liteをファインチューニングしたモデルです。", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "deberta-v2-base-japanese-finetuned-ner" + "score": 1.7954626862455887e-06, + "project_name": "japanese-sexual-moderation-v2" }, { - "url": "https://huggingface.co/Formzu/bart-base-japanese", + "url": "https://huggingface.co/minkhantycc/translation-en-ja", "downloads": 12, - "description": "bart-base-japaneseThis model is converted from the original Japanese BART Pretrained model released by Kyoto University.", + "description": "This model is the fine-tuned version of Helsinki-NLP/opus-mt-ja-en on bsd_ja_en dataset.", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "bart-base-japanese" + "score": 1.7954626862455887e-06, + "project_name": "translation-en-ja" }, { - "url": "https://huggingface.co/KoichiYasuoka/deberta-large-japanese-unidic-ud-head", + "url": "https://huggingface.co/dummy-foo/ChatGLM3-Japanese", "downloads": 12, - "description": "deberta-large-japanese-unidic-ud-headModel", + "description": "ChatGLM3-6B是一个中英双语大模型，本项目为ChatGLM3-6B加入日文能力。", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "deberta-large-japanese-unidic-ud-head" + "score": 1.7954626862455887e-06, + "project_name": "ChatGLM3-Japanese" }, { - "url": "https://huggingface.co/KoichiYasuoka/deberta-small-japanese-upos", + "url": "https://huggingface.co/line-corporation/japanese-large-lm-1.7b-instruction-sft-4bit-128g-actorder_False", "downloads": 12, - "description": "deberta-small-japanese-uposModel DescriptionThis is a DeBERTa(V2) model pre-trained on 青空文庫 texts for POS-tagging and dependency-parsing, derived from deberta-small-japanese-aozora.", + "description": "japanese-large-lm-1.7b-instruction-sft-4bit-128g-actorder_FalseThis repository provides a 1.7B parameters Japanese language quantized model, fine-tuned and trained by LINE Corporation.", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "deberta-small-japanese-upos" + "score": 1.7954626862455887e-06, + "project_name": "japanese-large-lm-1.7b-instruction-sft-4bit-128g-actorder_False" }, { - "url": "https://huggingface.co/KoichiYasuoka/roberta-small-japanese-char-luw-upos", + "url": "https://huggingface.co/ganchengguang/Yoko-7B-Japanese-v0", "downloads": 12, - "description": "roberta-small-japanese-char-luw-uposModel", + "description": "This model is traned with guanaco dataset.", "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "roberta-small-japanese-char-luw-upos" + "score": 1.7954626862455887e-06, + "project_name": "Yoko-7B-Japanese-v0" }, { - "url": "https://huggingface.co/KoichiYasuoka/bert-base-japanese-luw-upos", + "url": "https://huggingface.co/hitachi-nlp/bert-base-japanese_jumanpp-wordpiece", "downloads": 12, - "description": "bert-base-japanese-luw-uposModel", - "source": "Hugging Face", - "score": 1.6676744516026283e-06, - "project_name": "bert-base-japanese-luw-upos" - }, - { - "url": "https://huggingface.co/Helsinki-NLP/opus-mt-ja-bg", - "downloads": 11, - "description": "jpn-bulsource group: Japanesetarget group: BulgarianOPUS readme: jpn-bulmodel: transformer-alignsource language(s): jpn jpn_Hani jpn_Hira jpn_Kanatarget language(s): bulmodel: transformer-alignpre-processing: normalization + SentencePiece (spm32k,spm32k)", + "description": "Japanese BERT-base (Juman++ + WordPiece)How to load the tokenizerPlease download the dictionary file for Juman++ +", "source": "Hugging Face", - "score": 1.5287015806357426e-06, - "project_name": "opus-mt-ja-bg" + "score": 1.7954626862455887e-06, + "project_name": "bert-base-japanese_jumanpp-wordpiece" }, { - "url": "https://huggingface.co/KoichiYasuoka/roberta-small-japanese-aozora-char", - "downloads": 11, - "description": "roberta-small-japanese-aozora-charModel DescriptionThis is a RoBERTa model pre-trained on 青空文庫 texts with character tokenizer.", + "url": "https://huggingface.co/Mizuiro-sakura/bert-large-japanese-v2-finetuned-ner", + "downloads": 12, + "description": "このモデルはcl-tohoku/bert-large-japanese-v2をファインチューニングして、固有表現抽出（NER）に用いれるようにしたものです。", "source": "Hugging Face", - "score": 1.5287015806357426e-06, - "project_name": "roberta-small-japanese-aozora-char" + "score": 1.7954626862455887e-06, + "project_name": "bert-large-japanese-v2-finetuned-ner" }, { - "url": "https://huggingface.co/KoichiYasuoka/deberta-large-japanese-unidic-luw-upos", - "downloads": 11, - "description": "deberta-large-japanese-unidic-luw-uposModel", + "url": "https://huggingface.co/lorenzoncina/whisper-small-ja", + "downloads": 12, + "description": "Whisper Small JA - Lorenzo ConcinaThis model is a fine-tuned version of [SVJ Japanese dataset](https://huggingface.co/SVJ Japanese dataset) on the Common Voice 11.0 dataset.", "source": "Hugging Face", - "score": 1.5287015806357426e-06, - "project_name": "deberta-large-japanese-unidic-luw-upos" + "score": 1.7954626862455887e-06, + "project_name": "whisper-small-ja" }, { - "url": "https://huggingface.co/KoichiYasuoka/deberta-base-japanese-unidic-luw-upos", - "downloads": 11, - "description": "deberta-base-japanese-unidic-luw-uposModel", + "url": "https://huggingface.co/Language-Media-Lab/byt5-small-ain-jpn-mt", + "downloads": 12, + "description": "Byt5-small-ain-jpn-mt is a machine translation model pretrained with Google's ByT5-small and fine-tuned on bilingual datasets crawled from the Web.", "source": "Hugging Face", - "score": 1.5287015806357426e-06, - "project_name": "deberta-base-japanese-unidic-luw-upos" + "score": 1.7954626862455887e-06, + "project_name": "byt5-small-ain-jpn-mt" }, { - "url": "https://huggingface.co/KoichiYasuoka/deberta-large-japanese-luw-upos", - "downloads": 11, - "description": "deberta-large-japanese-luw-uposModel", + "url": "https://huggingface.co/astremo/friendly_JA", + "downloads": 12, + "description": "friendly_JA-Model　(T5 fine-tuned model)MT model trained using the friendly_JA Corpus attempting to make Japanese easier/more accessible to occidental people by using the Latin/English derived katakana lexicon instead of the standard Sino-Japanese lexiconExamplesinputoutput最適化を応用した機械翻訳モデルは高精度だオプティマイゼーションを応用したマシントランスレーションモデルは高いアキュラシーだ彼は架空の世界に住んでいる彼はイマジナリー世界に住んでいる新型コロナウイルスに感染してしまったコロナウイルスにかかってしまった深層学習は難しいディープラーニングはむずかしい新たな概念を紹介する新しいコンセプトを紹介する津波の警報が流れたツナミのアラートが流れた南海トラフの災害は震源地による南海トラフのディザスターはエピセンターによる息子は際どい内容の本を", "source": "Hugging Face", - "score": 1.5287015806357426e-06, - "project_name": "deberta-large-japanese-luw-upos" + "score": 1.7954626862455887e-06, + "project_name": "friendly_JA" }, { - "url": "https://huggingface.co/thefrigidliquidation/nllb-200-distilled-1.3B-bookworm", - "downloads": 11, - "description": "NLLB-200 1.3B fine-tuned on Ascendance of a BookwormThis model was fine-tuned on Ascendance of a Bookworm to translate the web novel in Japanese to English.", + "url": "https://huggingface.co/cinmodel/electra-small-japanese-discriminator", + "downloads": 12, + "description": "Japanese ELECTRA-smallWe provide a Japanese ELECTRA-Small model, as described in ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators.", "source": "Hugging Face", - "score": 1.5287015806357426e-06, - "project_name": "nllb-200-distilled-1.3B-bookworm" + "score": 1.7954626862455887e-06, + "project_name": "electra-small-japanese-discriminator" }, { - "url": "https://huggingface.co/KoichiYasuoka/roberta-large-japanese-aozora-ud-goeswith", + "url": "https://huggingface.co/KoichiYasuoka/roberta-small-japanese-char-luw-upos", "downloads": 11, - "description": "roberta-large-japanese-aozora-ud-goeswithModel DescriptionThis is a RoBERTa model pretrained on 青空文庫 texts for POS-tagging and dependency-parsing (using goeswith for subwords), derived from roberta-large-japanese-aozora and UD_Japanese-GSDLUW.How to Useclass UDgoeswith(object):def __init__(self,bert):", + "description": "roberta-small-japanese-char-luw-uposModel", "source": "Hugging Face", - "score": 1.5287015806357426e-06, - "project_name": "roberta-large-japanese-aozora-ud-goeswith" + "score": 1.6458407957251231e-06, + "project_name": "roberta-small-japanese-char-luw-upos" }, { - "url": "https://huggingface.co/hitachi-nlp/bert-base-japanese_nothing-unigram", + "url": "https://huggingface.co/KoichiYasuoka/roberta-large-japanese-char-luw-upos", "downloads": 11, - "description": "Japanese BERT-base (Nothing + Unigram)How to load the tokenizerPlease download the dictionary file for Nothing + Unigram from our GitHub repository.", + "description": "roberta-large-japanese-char-luw-uposModel", "source": "Hugging Face", - "score": 1.5287015806357426e-06, - "project_name": "bert-base-japanese_nothing-unigram" + "score": 1.6458407957251231e-06, + "project_name": "roberta-large-japanese-char-luw-upos" }, { - "url": "https://huggingface.co/hitachi-nlp/bert-base-japanese_vaporetto-bpe", + "url": "https://huggingface.co/knok/japanese-distilgpt2", "downloads": 11, - "description": "Japanese BERT-base (Vaporetto + BPE)How to load the tokenizerPlease download the dictionary file for Vaporetto + BPE from our GitHub repository.", + "description": "日本語 gpt2 蒸留モデルこのモデルはrinna/japanese-gpt2-meduimを教師として蒸留したものです。", "source": "Hugging Face", - "score": 1.5287015806357426e-06, - "project_name": "bert-base-japanese_vaporetto-bpe" + "score": 1.6458407957251231e-06, + "project_name": "japanese-distilgpt2" }, { - "url": "https://huggingface.co/hitachi-nlp/bert-base-japanese_sudachi-wordpiece", + "url": "https://huggingface.co/hitachi-nlp/bert-base-japanese_jumanpp-unigram", "downloads": 11, - "description": "Japanese BERT-base (Sudachi + WordPiece)How to load the tokenizerPlease download the dictionary file for Sudachi + WordPiece from our GitHub repository.", + "description": "Japanese BERT-base (Juman++ + Unigram)How to load the tokenizerPlease download the dictionary file for Juman++ + Unigram from our GitHub repository.", "source": "Hugging Face", - "score": 1.5287015806357426e-06, - "project_name": "bert-base-japanese_sudachi-wordpiece" + "score": 1.6458407957251231e-06, + "project_name": "bert-base-japanese_jumanpp-unigram" }, { - "url": "https://huggingface.co/hitachi-nlp/bert-base-japanese_sudachi-bpe", + "url": "https://huggingface.co/hitachi-nlp/bert-base-japanese_mecab-unigram", "downloads": 11, - "description": "Japanese BERT-base (Sudachi + BPE)How to load the tokenizerPlease download the dictionary file for Sudachi + BPE from our GitHub repository.", + "description": "Japanese BERT-base (MeCab + Unigram)How to load the tokenizerPlease download the dictionary file for MeCab + Unigram from our GitHub repository.", "source": "Hugging Face", - "score": 1.5287015806357426e-06, - "project_name": "bert-base-japanese_sudachi-bpe" + "score": 1.6458407957251231e-06, + "project_name": "bert-base-japanese_mecab-unigram" }, { - "url": "https://huggingface.co/if001/llama2_ja_small_instruct", + "url": "https://huggingface.co/hitachi-nlp/bert-base-japanese_mecab-bpe", "downloads": 11, - "description": "日本語でtrainingしたllama2をinstruction用のデータセットでsftしたものになりますbase:https://huggingface.co/if001/llama2_ja_smalltrainingは以下のscript参照https://github.com/Lightning-AI/lit-gpt/tree/mainusefrom transformers import AutoTokenizer, AutoModelForCausalLMtokenizer = AutoTokenizer.from_pretrained(\"if001/sentencepiece_ja\", trust_remote_code=True)model = AutoModelForCausalLM.from_pretrained(\"if001/llama2_ja_small\")", + "description": "Japanese BERT-base (MeCab + BPE)How to load the tokenizerPlease download the dictionary file for MeCab + BPE from our GitHub repository.", "source": "Hugging Face", - "score": 1.5287015806357426e-06, - "project_name": "llama2_ja_small_instruct" + "score": 1.6458407957251231e-06, + "project_name": "bert-base-japanese_mecab-bpe" }, { - "url": "https://huggingface.co/GralchemOz/Qwen1.5-14B-vntl-jp2zh-4.5bpw-h6-exl2", + "url": "https://huggingface.co/eepj/wstcg-mt-ja-en", "downloads": 11, - "description": "This model is a merged version of qwen-14b-vntl and Qwen1.5-14B-Chat, aiming for the translation of Japanese context into Chinese.", + "description": "WS TCG Card Text TranslatorA Japanese-English machine translation model specifically trained for translating card text from the Weiss Schwarz (WS) Trading Card Game, fine-tuned on Helsinki-NLP/opus-mt-ja-en.", "source": "Hugging Face", - "score": 1.5287015806357426e-06, - "project_name": "Qwen1.5-14B-vntl-jp2zh-4.5bpw-h6-exl2" + "score": 1.6458407957251231e-06, + "project_name": "wstcg-mt-ja-en" }, { - "url": "https://huggingface.co/dummy-foo/ChatGLM3-Japanese", + "url": "https://huggingface.co/knosing/japanese_ner_model", "downloads": 11, - "description": "ChatGLM3-6B是一个中英双语大模型，本项目为ChatGLM3-6B加入日文能力。", + "description": "Model DescriptionThis model is a fine-tuned version of the tohoku-nlp/bert-base-japanese-v3, specifically optimized for Named Entity Recognition (NER) tasks.", "source": "Hugging Face", - "score": 1.5287015806357426e-06, - "project_name": "ChatGLM3-Japanese" + "score": 1.6458407957251231e-06, + "project_name": "japanese_ner_model" }, { - "url": "https://huggingface.co/Kendamarron/Tokara-0.5B-Chat-v0.1", + "url": "https://huggingface.co/Local-Novel-LLM-project/Vecteus-Poet", "downloads": 11, - "description": "モデルについてQwen/Qwen1.5-0.5Bを日英データ5Bトークンで継続事前学習したTokara-0.5B-v0.1にchat vectorで対話能力を加えたモデルになります。", + "description": "Our ModelsVecteusNinja-v1Ninja-v1-NSFWNinja-v1-128kNinja-v1-NSFW-128kThis is a prototype of Vecteus-v1Model Card for VecTeus-PoetThe Mistral-7B--based Large Language Model (LLM) is an noveldataset fine-tuned version of the Mistral-7B-v0.1VecTeus has the following changes compared to Mistral-7B-v0.1.Achieving both high quality Japanese and English generationCan be generated NSFWMemory ability that does not forget even after long-context generationThis model was created with the help of GPUs from the first Lo", "source": "Hugging Face", - "score": 1.5287015806357426e-06, - "project_name": "Tokara-0.5B-Chat-v0.1" + "score": 1.6458407957251231e-06, + "project_name": "Vecteus-Poet" }, { "url": "https://huggingface.co/ohwi/japanese-stablelm-instruct-gamma-7b-dpo-uf-v1", "downloads": 11, "description": "Japanese Stable LM Instruct Gamma 7B + DPOModel", "source": "Hugging Face", - "score": 1.5287015806357426e-06, + "score": 1.6458407957251231e-06, "project_name": "japanese-stablelm-instruct-gamma-7b-dpo-uf-v1" }, { - "url": "https://huggingface.co/NilanE/tinyllama-en_ja-translation-v2", - "downloads": 11, - "description": "In-progess long-context Japanese-English translation model based on tinyllama.", - "source": "Hugging Face", - "score": 1.5287015806357426e-06, - "project_name": "tinyllama-en_ja-translation-v2" - }, - { - "url": "https://huggingface.co/TheBloke/japanese-stablelm-instruct-gamma-7B-AWQ", + "url": "https://huggingface.co/AIgroup-CVM-utokyohospital/Llama-2-70b-chat-4bit-japanese", "downloads": 11, - "description": "Chat & support: TheBloke's Discord serverWant to contribute?", + "description": "AIgroup-CVM-utokyohospital/Llama-2-70b-chat-4bit-japaneseThis model is Llama-2-Chat 70B fine-tuned with a part of the Japanese instruction dataset named izumi-lab/llm-japanese-dataset.", "source": "Hugging Face", - "score": 1.5287015806357426e-06, - "project_name": "japanese-stablelm-instruct-gamma-7B-AWQ" + "score": 1.6458407957251231e-06, + "project_name": "Llama-2-70b-chat-4bit-japanese" }, { - "url": "https://huggingface.co/hitachi-nlp/bert-base-japanese_mecab-unigram", + "url": "https://huggingface.co/hitachi-nlp/bert-base-japanese_nothing-bpe", "downloads": 11, - "description": "Japanese BERT-base (MeCab + Unigram)How to load the tokenizerPlease download the dictionary file for MeCab + Unigram from our GitHub repository.", + "description": "Japanese BERT-base (Nothing + BPE)How to load the tokenizerPlease download the dictionary file for Nothing + BPE from our GitHub repository.", "source": "Hugging Face", - "score": 1.5287015806357426e-06, - "project_name": "bert-base-japanese_mecab-unigram" + "score": 1.6458407957251231e-06, + "project_name": "bert-base-japanese_nothing-bpe" }, { - "url": "https://huggingface.co/hitachi-nlp/bert-base-japanese_mecab-bpe", + "url": "https://huggingface.co/hitachi-nlp/bert-base-japanese_vaporetto-wordpiece", "downloads": 11, - "description": "Japanese BERT-base (MeCab + BPE)How to load the tokenizerPlease download the dictionary file for MeCab + BPE from our GitHub repository.", + "description": "Japanese BERT-base (Vaporetto + WordPiece)How to load the tokenizerPlease download the dictionary file for Vaporetto + WordPiece from our GitHub repository.", "source": "Hugging Face", - "score": 1.5287015806357426e-06, - "project_name": "bert-base-japanese_mecab-bpe" + "score": 1.6458407957251231e-06, + "project_name": "bert-base-japanese_vaporetto-wordpiece" }, { - "url": "https://huggingface.co/Mizuiro-sakura/deberta-v2-tiny-japanese-finetuned-QA", + "url": "https://huggingface.co/hitachi-nlp/bert-base-japanese_vaporetto-bpe", "downloads": 11, - "description": "このモデルはdeberta-v2-tiny-japaneseをファインチューニングしてQAタスクに用いれるようにしたものです。", + "description": "Japanese BERT-base (Vaporetto + BPE)How to load the tokenizerPlease download the dictionary file for Vaporetto + BPE from our GitHub repository.", "source": "Hugging Face", - "score": 1.5287015806357426e-06, - "project_name": "deberta-v2-tiny-japanese-finetuned-QA" + "score": 1.6458407957251231e-06, + "project_name": "bert-base-japanese_vaporetto-bpe" }, { - "url": "https://huggingface.co/KoichiYasuoka/deberta-base-japanese-wikipedia", + "url": "https://huggingface.co/hitachi-nlp/bert-base-japanese_sudachi-wordpiece", "downloads": 11, - "description": "deberta-base-japanese-wikipediaModel DescriptionThis is a DeBERTa(V2) model pre-trained on Japanese Wikipedia and 青空文庫 texts.", + "description": "Japanese BERT-base (Sudachi + WordPiece)How to load the tokenizerPlease download the dictionary file for Sudachi + WordPiece from our GitHub repository.", "source": "Hugging Face", - "score": 1.5287015806357426e-06, - "project_name": "deberta-base-japanese-wikipedia" + "score": 1.6458407957251231e-06, + "project_name": "bert-base-japanese_sudachi-wordpiece" }, { - "url": "https://huggingface.co/KoichiYasuoka/deberta-base-japanese-luw-upos", + "url": "https://huggingface.co/akiFQC/japanese-dialogpt-small-aozora", "downloads": 11, - "description": "deberta-base-japanese-luw-uposModel", + "description": "Japanese DialoGPT trained with Aozora(ja) 青空文庫のセリフで学習した日本語のDialoGPT Smallです(en) Japanese DialoGPT Small trained on Aozora Bunko.", "source": "Hugging Face", - "score": 1.5287015806357426e-06, - "project_name": "deberta-base-japanese-luw-upos" + "score": 1.6458407957251231e-06, + "project_name": "japanese-dialogpt-small-aozora" }, { - "url": "https://huggingface.co/KoichiYasuoka/deberta-small-japanese-luw-upos", + "url": "https://huggingface.co/Formzu/bart-base-japanese", "downloads": 11, - "description": "deberta-small-japanese-luw-uposModel", + "description": "bart-base-japaneseThis model is converted from the original Japanese BART Pretrained model released by Kyoto University.", "source": "Hugging Face", - "score": 1.5287015806357426e-06, - "project_name": "deberta-small-japanese-luw-upos" + "score": 1.6458407957251231e-06, + "project_name": "bart-base-japanese" }, { - "url": "https://huggingface.co/Language-Media-Lab/byt5-small-ain-jpn-mt", + "url": "https://huggingface.co/Formzu/bert-base-japanese-jsnli", "downloads": 11, - "description": "Byt5-small-ain-jpn-mt is a machine translation model pretrained with Google's ByT5-small and fine-tuned on bilingual datasets crawled from the Web.", + "description": "bert-base-japanese-jsnliThis model is a fine-tuned version of cl-tohoku/bert-base-japanese-v2 on the JSNLI dataset.", "source": "Hugging Face", - "score": 1.5287015806357426e-06, - "project_name": "byt5-small-ain-jpn-mt" + "score": 1.6458407957251231e-06, + "project_name": "bert-base-japanese-jsnli" }, { "url": "https://huggingface.co/datasets/shunk031/JGLUE", - "downloads": 29844, - "description": "Please feel free to open an issue or pull request.", - "source": "Hugging Face", - "score": 0.004147506361135736, - "project_name": "JGLUE" - }, - { - "url": "https://huggingface.co/datasets/sbintuitions/JMTEB", - "downloads": 11528, - "description": "JMTEB:", - "source": "Hugging Face", - "score": 0.0016020792565062583, - "project_name": "JMTEB" - }, - { - "url": "https://huggingface.co/datasets/elyza/ELYZA-tasks-100", - "downloads": 3912, - "description": "ELYZA-tasks-100: 日本語instructionモデル評価データセットData Description本データセットはinstruction-tuningを行ったモデルの評価用データセットです。", - "source": "Hugging Face", - "score": 0.0005436618712224568, - "project_name": "ELYZA-tasks-100" - }, - { - "url": "https://huggingface.co/datasets/nlp-waseda/JMMLU", - "downloads": 3765, - "description": "JMMLUJapanese Massive Multitask Language Understanding BenchmarkJMMLU is a four-choice question set consisting of Japanese-translated questions of a portion of MMLU (Paper, Github) (Translated questions) and questions based on unique Japanese cultural context (Japanese questions).", - "source": "Hugging Face", - "score": 0.0005232328591903246, - "project_name": "JMMLU" - }, - { - "url": "https://huggingface.co/datasets/kunishou/databricks-dolly-15k-ja", - "downloads": 2809, - "description": "This dataset was created by automatically translating \"databricks-dolly-15k\" into Japanese.", + "downloads": 29986, + "description": "Please feel free to open an issue or pull request.", "source": "Hugging Face", - "score": 0.0003903747945459819, - "project_name": "databricks-dolly-15k-ja" + "score": 0.0044865620091466855, + "project_name": "JGLUE" }, { - "url": "https://huggingface.co/datasets/llm-jp/hh-rlhf-12k-ja", - "downloads": 2590, - "description": "hh-rlhf-12k-jaThis repository provides a human preference dataset developed by LLM-jp, a collaborative project launched in Japan.", + "url": "https://huggingface.co/datasets/sbintuitions/JMTEB", + "downloads": 11753, + "description": "JMTEB:", "source": "Hugging Face", - "score": 0.00035993973580423393, - "project_name": "hh-rlhf-12k-ja" + "score": 0.0017585060792870337, + "project_name": "JMTEB" }, { - "url": "https://huggingface.co/datasets/mkshing/xlsum_ja", - "downloads": 1675, - "description": "This is the filtered Japanese subset of XL-Sum followed by PaLM 2filters15-gram overlap* code: https://gist.github.com/mkshing/d6371cbfdd50d4f352cee247fd4dd86anumber of examplestrain: 4215 (before: 7113)validation: 758 (before: 889)test: 766 (before: 889)", + "url": "https://huggingface.co/datasets/elyza/ELYZA-tasks-100", + "downloads": 3296, + "description": "ELYZA-tasks-100: 日本語instructionモデル評価データセットData Description本データセットはinstruction-tuningを行ったモデルの評価用データセットです。", "source": "Hugging Face", - "score": 0.0002327795588695335, - "project_name": "xlsum_ja" + "score": 0.0004931537511554551, + "project_name": "ELYZA-tasks-100" }, { - "url": "https://huggingface.co/datasets/llm-jp/databricks-dolly-15k-ja", - "downloads": 1672, - "description": "databricks-dolly-15k-jaThis repository provides an instruction tuning dataset developed by LLM-jp, a collaborative project launched in Japan.", + "url": "https://huggingface.co/datasets/kunishou/databricks-dolly-15k-ja", + "downloads": 3076, + "description": "This dataset was created by automatically translating \"databricks-dolly-15k\" into Japanese.", "source": "Hugging Face", - "score": 0.00023236264025663285, + "score": 0.0004602369352409526, "project_name": "databricks-dolly-15k-ja" }, { - "url": "https://huggingface.co/datasets/range3/wiki40b-ja", - "downloads": 1380, - "description": "range3/wiki40b-jaThis dataset consists of three parquet files from the wiki40b dataset with only Japanese data extracted.", + "url": "https://huggingface.co/datasets/nlp-waseda/JMMLU", + "downloads": 2564, + "description": "JMMLUJapanese Massive Multitask Language Understanding BenchmarkJMMLU is a four-choice question set consisting of Japanese-translated questions of a portion of MMLU (Paper, Github) (Translated questions) and questions based on unique Japanese cultural context (Japanese questions).", "source": "Hugging Face", - "score": 0.00019178256193430225, - "project_name": "wiki40b-ja" + "score": 0.00038363052729447413, + "project_name": "JMMLU" }, { "url": "https://huggingface.co/datasets/cl-nagoya/auto-wiki-qa", - "downloads": 1359, + "downloads": 1845, "description": "AutoWikiQA東工大が公開しているSwallow-MXを用いて、Wikipedia中のテキストを入力として「質問(query)」と「回答(answer)」を生成し、生成された質問と回答についてフィルタリングを行ったデータセットです。", "source": "Hugging Face", - "score": 0.00018886413164399764, + "score": 0.0002760523880102593, "project_name": "auto-wiki-qa" }, + { + "url": "https://huggingface.co/datasets/mkshing/xlsum_ja", + "downloads": 1519, + "description": "This is the filtered Japanese subset of XL-Sum followed by PaLM 2filters15-gram overlap* code: https://gist.github.com/mkshing/d6371cbfdd50d4f352cee247fd4dd86anumber of examplestrain: 4215 (before: 7113)validation: 758 (before: 889)test: 766 (before: 889)", + "source": "Hugging Face", + "score": 0.00022727565170058744, + "project_name": "xlsum_ja" + }, { "url": "https://huggingface.co/datasets/kumapo/JAQKET", - "downloads": 1346, + "downloads": 1451, "description": "作成するデータセットは，既存研究 [7] に倣い，Wikipedia2 の記事名を答えとした，日本語のオープンドメイン QA タスクのデータセットである.", "source": "Hugging Face", - "score": 0.00018705748432142812, + "score": 0.00021710136314519578, "project_name": "JAQKET" }, { "url": "https://huggingface.co/datasets/llm-book/wrime-sentiment", - "downloads": 1158, + "downloads": 1165, "description": "GitHub リポジトリ ids-cv/wrime で公開されているデータセットを利用しています。", "source": "Hugging Face", - "score": 0.0001609305845796536, + "score": 0.00017430950245634258, "project_name": "wrime-sentiment" }, + { + "url": "https://huggingface.co/datasets/shunk031/jsnli", + "downloads": 1116, + "description": "Dataset PreprocessingSupported Tasks and LeaderboardsLanguages注釈はすべて日本語を主要言語としています。", + "source": "Hugging Face", + "score": 0.00016697802982083975, + "project_name": "jsnli" + }, { "url": "https://huggingface.co/datasets/llm-book/ner-wikipedia-dataset", - "downloads": 784, + "downloads": 1090, "description": "Githubリポジトリstockmarkteam/ner-wikipedia-datasetで公開されているデータセットを利用しています。", "source": "Hugging Face", - "score": 0.00010895473083803838, + "score": 0.00016308786066730765, "project_name": "ner-wikipedia-dataset" }, + { + "url": "https://huggingface.co/datasets/llm-jp/hh-rlhf-12k-ja", + "downloads": 976, + "description": "hh-rlhf-12k-jaThis repository provides a human preference dataset developed by LLM-jp, a collaborative project launched in Japan.", + "source": "Hugging Face", + "score": 0.00014603096514797454, + "project_name": "hh-rlhf-12k-ja" + }, { "url": "https://huggingface.co/datasets/turing-motors/LLaVA-Instruct-150K-JA", "downloads": 725, "description": "Dataset DetailsDataset Type:Japanese LLaVA Instruct 150K is a localized version of the original LLaVA Visual Instruct 150K dataset.", "source": "Hugging Face", - "score": 0.00010075533145099212, + "score": 0.00010847587062733766, "project_name": "LLaVA-Instruct-150K-JA" }, { "url": "https://huggingface.co/datasets/hatakeyama-llm-team/AutoGeneratedJapaneseQA", - "downloads": 662, + "downloads": 697, "description": "自動生成Q&A種々のデータソースから､MaziyarPanahi/Mixtral-8x22B-Instruct-v0.1-GGUFを使ってQ&Aを自動生成したものです｡CC-BY系またはApatch-2.0のデータソースを改変して生成しています。", "source": "Hugging Face", - "score": 9.200004058007832e-05, + "score": 0.00010428645769276461, "project_name": "AutoGeneratedJapaneseQA" }, { "url": "https://huggingface.co/datasets/neulab/odex", - "downloads": 591, + "downloads": 628, "description": "ODEX is an Open-Domain EXecution-based NL-to-Code generation data benchmark.", "source": "Hugging Face", - "score": 8.213296674142944e-05, + "score": 9.396254724685248e-05, "project_name": "odex" }, { - "url": "https://huggingface.co/datasets/seungwon929/Ja-miracl", - "downloads": 561, - "description": "Ja-miraclThis dataset represents a conversion of the Japanese (Ja) section from the miracl dataset into the BeIR format, making it compatible for use with mteb.", - "source": "Hugging Face", - "score": 7.796378061242287e-05, - "project_name": "Ja-miracl" - }, - { - "url": "https://huggingface.co/datasets/shunk031/jsnli", - "downloads": 538, - "description": "Dataset PreprocessingSupported Tasks and LeaderboardsLanguages注釈はすべて日本語を主要言語としています。", + "url": "https://huggingface.co/datasets/llm-book/jawiki-sentences", + "downloads": 507, + "description": "GitHub リポジトリ singletongue/wikipedia-utils で公開されているデータセットを利用しています。", "source": "Hugging Face", - "score": 7.47674045801845e-05, - "project_name": "jsnli" + "score": 7.585829849387613e-05, + "project_name": "jawiki-sentences" }, { "url": "https://huggingface.co/datasets/kunishou/J-ResearchCorpus", - "downloads": 473, + "downloads": 465, "description": "J-ResearchCorpusUpdate:2024/3/16言語処理学会第30回年次大会(NLP2024)を含む、論文 1,343 本のデータを追加2024/2/25言語処理学会誌「自然言語処理」のうち CC-BY-4.0 で公開されている論文 360 本のデータを追加概要CC-BY-* ライセンスで公開されている日本語論文や学会誌等から抜粋した高品質なテキストのデータセットです。", "source": "Hugging Face", - "score": 6.573416796733693e-05, + "score": 6.957417909201656e-05, "project_name": "J-ResearchCorpus" }, { - "url": "https://huggingface.co/datasets/llm-book/jawiki-sentences", - "downloads": 467, - "description": "GitHub リポジトリ singletongue/wikipedia-utils で公開されているデータセットを利用しています。", + "url": "https://huggingface.co/datasets/hotchpotch/JQaRA", + "downloads": 462, + "description": "JQaRA : Japanese Question Answering with Retrieval Augmentation - 検索拡張(RAG)評価のための日本語 Q&A データセット高性能な LLM の台頭に伴い、LLM を用いた質疑応答のユースケースが増加しています。", "source": "Hugging Face", - "score": 6.490033074153562e-05, - "project_name": "jawiki-sentences" + "score": 6.912531342045517e-05, + "project_name": "JQaRA" }, { "url": "https://huggingface.co/datasets/izumi-lab/llm-japanese-dataset", - "downloads": 430, + "downloads": 443, "description": "llm-japanese-datasetLLM構築用の日本語インストラクション(チャット)データセット主に，英語で構築されたLLMモデルなどに対して，チャット(Instruction)応答タスクに関してLoRAなどでチューニングするために使用できます．", "source": "Hugging Face", - "score": 5.975833451576084e-05, + "score": 6.628249750056632e-05, "project_name": "llm-japanese-dataset" }, { - "url": "https://huggingface.co/datasets/hotchpotch/JQaRA", - "downloads": 425, - "description": "JQaRA : Japanese Question Answering with Retrieval Augmentation - 検索拡張(RAG)評価のための日本語 Q&A データセット高性能な LLM の台頭に伴い、LLM を用いた質疑応答のユースケースが増加しています。", + "url": "https://huggingface.co/datasets/llm-book/livedoor-news-corpus", + "downloads": 419, + "description": "オリジナルのサイトと同じものを使用しています。", "source": "Hugging Face", - "score": 5.906347016092641e-05, - "project_name": "JQaRA" + "score": 6.269157212807514e-05, + "project_name": "livedoor-news-corpus" + }, + { + "url": "https://huggingface.co/datasets/range3/wiki40b-ja", + "downloads": 384, + "description": "range3/wiki40b-jaThis dataset consists of three parquet files from the wiki40b dataset with only Japanese data extracted.", + "source": "Hugging Face", + "score": 5.745480595985884e-05, + "project_name": "wiki40b-ja" }, { "url": "https://huggingface.co/datasets/kogi-jwu/jhumaneval", - "downloads": 409, + "downloads": 376, "description": "LLM のコード生成能力の標準ベンチマーク HumanEval の日本語翻訳版です。", "source": "Hugging Face", - "score": 5.6839904225456246e-05, + "score": 5.625783083569512e-05, "project_name": "jhumaneval" }, { - "url": "https://huggingface.co/datasets/globis-university/aozorabunko-clean", - "downloads": 378, - "description": "OverviewThis dataset provides a convenient and user-friendly format of data from Aozora Bunko (青空文庫), a website that compiles public-domain books in Japan, ideal for Machine Learning applications.", + "url": "https://huggingface.co/datasets/llm-book/aio-retriever", + "downloads": 371, + "description": "GitHub リポジトリ cl-tohoku/quiz-datasets で公開されているデータセットを利用しています。", "source": "Hugging Face", - "score": 5.2531745225482786e-05, - "project_name": "aozorabunko-clean" + "score": 5.550972138309279e-05, + "project_name": "aio-retriever" }, { - "url": "https://huggingface.co/datasets/llm-book/livedoor-news-corpus", - "downloads": 358, - "description": "オリジナルのサイトと同じものを使用しています。", + "url": "https://huggingface.co/datasets/globis-university/aozorabunko-clean", + "downloads": 361, + "description": "OverviewThis dataset provides a convenient and user-friendly format of data from Aozora Bunko (青空文庫), a website that compiles public-domain books in Japan, ideal for Machine Learning applications.", "source": "Hugging Face", - "score": 4.9752287806145074e-05, - "project_name": "livedoor-news-corpus" + "score": 5.401350247788813e-05, + "project_name": "aozorabunko-clean" }, { - "url": "https://huggingface.co/datasets/llm-jp/oasst1-21k-ja", + "url": "https://huggingface.co/datasets/yuzuai/rakuda-questions", "downloads": 336, - "description": "oasst1-21k-jaThis repository provides an instruction tuning dataset developed by LLM-jp, a collaborative project launched in Japan.", + "description": "Rakuda - Questions for Japanese modelsRepository:", "source": "Hugging Face", - "score": 4.669488464487359e-05, - "project_name": "oasst1-21k-ja" + "score": 5.0272955214876485e-05, + "project_name": "rakuda-questions" }, { - "url": "https://huggingface.co/datasets/saldra/sakura_japanese_dataset", - "downloads": 287, - "description": "Sakura_dataset商用利用可能な超小規模高品質日本語データセット。", + "url": "https://huggingface.co/datasets/llm-jp/databricks-dolly-15k-ja", + "downloads": 310, + "description": "databricks-dolly-15k-jaThis repository provides an instruction tuning dataset developed by LLM-jp, a collaborative project launched in Japan.", "source": "Hugging Face", - "score": 3.988521396749619e-05, - "project_name": "sakura_japanese_dataset" + "score": 4.638278606134438e-05, + "project_name": "databricks-dolly-15k-ja" }, { - "url": "https://huggingface.co/datasets/hatakeyama-llm-team/AutoGeneratedJapaneseQA-other", - "downloads": 226, - "description": "自動生成Q&Aデータソースから､MaziyarPanahi/Mixtral-8x22B-Instruct-v0.1-GGUFを使ってQ&Aを自動生成したものです｡チームで作成したデータおよび「Common Crawlをもとに生成しています。 ", + "url": "https://huggingface.co/datasets/seungwon929/Ja-miracl", + "downloads": 306, + "description": "Ja-miraclThis dataset represents a conversion of the Japanese (Ja) section from the miracl dataset into the BeIR format, making it compatible for use with mteb.", "source": "Hugging Face", - "score": 3.140786883851617e-05, - "project_name": "AutoGeneratedJapaneseQA-other" + "score": 4.578429849926251e-05, + "project_name": "Ja-miracl" }, { - "url": "https://huggingface.co/datasets/llm-book/aio-retriever", - "downloads": 224, - "description": "GitHub リポジトリ cl-tohoku/quiz-datasets で公開されているデータセットを利用しています。", + "url": "https://huggingface.co/datasets/llm-jp/oasst1-21k-ja", + "downloads": 252, + "description": "oasst1-21k-jaThis repository provides an instruction tuning dataset developed by LLM-jp, a collaborative project launched in Japan.", "source": "Hugging Face", - "score": 3.1129923096582394e-05, - "project_name": "aio-retriever" + "score": 3.7704716411157364e-05, + "project_name": "oasst1-21k-ja" }, { - "url": "https://huggingface.co/datasets/yuzuai/rakuda-questions", - "downloads": 222, - "description": "Rakuda - Questions for Japanese modelsRepository:", + "url": "https://huggingface.co/datasets/saldra/sakura_japanese_dataset", + "downloads": 242, + "description": "Sakura_dataset商用利用可能な超小規模高品質日本語データセット。", "source": "Hugging Face", - "score": 3.085197735464862e-05, - "project_name": "rakuda-questions" + "score": 3.6208497505952704e-05, + "project_name": "sakura_japanese_dataset" }, { "url": "https://huggingface.co/datasets/creative-graphic-design/CAMERA", "downloads": 212, "description": "We hope that our dataset will be useful in research for realizing more advanced ad text generation models.", "source": "Hugging Face", - "score": 2.9462248644979765e-05, + "score": 3.171984079033873e-05, "project_name": "CAMERA" }, { - "url": "https://huggingface.co/datasets/DataPilot/databricks-dolly-15k-Nyan-ja", - "downloads": 166, - "description": "このデータセットはkunishou氏が公開している\"databricks-dolly-15k\"を日本語訳したkunishou/databricks-dolly-15k-jaデータセットの語尾をArrowPro-7B-KUJIRAを用いて「にゃん！", - "source": "Hugging Face", - "score": 2.3069496580503025e-05, - "project_name": "databricks-dolly-15k-Nyan-ja" - }, - { - "url": "https://huggingface.co/datasets/SkelterLabsInc/JaQuAD", - "downloads": 162, - "description": "JaQuAD is developed to provide a SQuAD-like QA dataset in Japanese.", + "url": "https://huggingface.co/datasets/range3/cc100-ja", + "downloads": 190, + "description": "range3/cc100-jaThis dataset consists of parquet files from the cc100 dataset with only the Japanese language extracted and sharded.", "source": "Hugging Face", - "score": 2.251360509663548e-05, - "project_name": "JaQuAD" + "score": 2.842815919888849e-05, + "project_name": "cc100-ja" }, { - "url": "https://huggingface.co/datasets/HachiML/alpaca_jp_python", - "downloads": 160, - "description": "alpaca_jp_pythonalpaca_jp_pythonは、Stanford Alpacaの手法mistralai/Mixtral-8x22B-Instruct-v0.1で作った合成データ(Synthetic data)です。", + "url": "https://huggingface.co/datasets/hatakeyama-llm-team/AutoGeneratedJapaneseQA-other", + "downloads": 175, + "description": "自動生成Q&Aデータソースから､MaziyarPanahi/Mixtral-8x22B-Instruct-v0.1-GGUFを使ってQ&Aを自動生成したものです｡チームで作成したデータおよび「Common Crawlをもとに生成しています。 ", "source": "Hugging Face", - "score": 2.223565935470171e-05, - "project_name": "alpaca_jp_python" + "score": 2.6183830841081504e-05, + "project_name": "AutoGeneratedJapaneseQA-other" }, { - "url": "https://huggingface.co/datasets/yongtae-jp/orca_dpo_pairs_ja", - "downloads": 158, - "description": "About this datasetThis dataset is a machine translation of the Intel/orca_dpo_pairs dataset with Palm 2 (prompt for translation is pasted below).", + "url": "https://huggingface.co/datasets/hpprc/jsick", + "downloads": 172, + "description": "Dataset.", "source": "Hugging Face", - "score": 2.1957713612767938e-05, - "project_name": "orca_dpo_pairs_ja" + "score": 2.5734965169520106e-05, + "project_name": "jsick" }, { "url": "https://huggingface.co/datasets/range3/wikipedia-ja-20230101", - "downloads": 156, + "downloads": 144, "description": "range3/wikipedia-ja-20230101This dataset consists of a parquet file from the wikipedia dataset with only Japanese data extracted.", "source": "Hugging Face", - "score": 2.1679767870834168e-05, + "score": 2.1545552234947066e-05, "project_name": "wikipedia-ja-20230101" }, - { - "url": "https://huggingface.co/datasets/nyanko7/danbooru2023", - "downloads": 139, - "description": "Danbooru2023:", - "source": "Hugging Face", - "score": 1.931722906439711e-05, - "project_name": "danbooru2023" - }, { "url": "https://huggingface.co/datasets/joujiboi/japanese-anime-speech", - "downloads": 133, + "downloads": 142, "description": "Japanese Anime Speech Dataset日本語はこちらjapanese-anime-speech is an audio-text dataset designed for the training of automatic speech recognition models.", "source": "Hugging Face", - "score": 1.8483391838595796e-05, + "score": 2.1246308453906134e-05, "project_name": "japanese-anime-speech" }, { - "url": "https://huggingface.co/datasets/kunishou/OpenMathInstruct-1-1.8m-ja", - "downloads": 131, - "description": "OpenMathInstruct-1 を日本語に自動翻訳した商用利用可能な180万件の指示チューニングデータセットになります。", + "url": "https://huggingface.co/datasets/SkelterLabsInc/JaQuAD", + "downloads": 141, + "description": "JaQuAD is developed to provide a SQuAD-like QA dataset in Japanese.", "source": "Hugging Face", - "score": 1.8205446096662024e-05, - "project_name": "OpenMathInstruct-1-1.8m-ja" + "score": 2.1096686563385668e-05, + "project_name": "JaQuAD" }, { - "url": "https://huggingface.co/datasets/baobab-trees/wikipedia-human-retrieval-ja", - "downloads": 129, - "description": "Japanese Wikipedia Human Retrieval datasetThis is a Japanese question answereing dataset with retrieval on Wikipedia articlesby trained human workers.", + "url": "https://huggingface.co/datasets/nyanko7/danbooru2023", + "downloads": 127, + "description": "Danbooru2023:", "source": "Hugging Face", - "score": 1.7927500354728254e-05, - "project_name": "wikipedia-human-retrieval-ja" + "score": 1.9001980096099148e-05, + "project_name": "danbooru2023" + }, + { + "url": "https://huggingface.co/datasets/kunishou/OpenMathInstruct-1-1.8m-ja", + "downloads": 126, + "description": "OpenMathInstruct-1 を日本語に自動翻訳���た商用利用可能な180万件の指示チューニングデータセットになります。", + "source": "Hugging Face", + "score": 1.8852358205578682e-05, + "project_name": "OpenMathInstruct-1-1.8m-ja" }, { "url": "https://huggingface.co/datasets/zan/lima-ja", "downloads": 126, "description": ", 2023) was trained on.", "source": "Hugging Face", - "score": 1.7510581741827595e-05, + "score": 1.8852358205578682e-05, "project_name": "lima-ja" }, { "url": "https://huggingface.co/datasets/fujiki/japanese_alpaca_data", - "downloads": 125, + "downloads": 123, "description": "[github].", "source": "Hugging Face", - "score": 1.737160887086071e-05, + "score": 1.8403492534017284e-05, "project_name": "japanese_alpaca_data" }, { - "url": "https://huggingface.co/datasets/kunishou/amenokaku-code-instruct", + "url": "https://huggingface.co/datasets/baobab-trees/wikipedia-human-retrieval-ja", "downloads": 123, - "description": "Amenokaku-Code-InstructUpdate:2023/12/27データセットに JaxTon , プロになるJava のコードデータ 180 レコードを追加しました。", - "source": "Hugging Face", - "score": 1.709366312892694e-05, - "project_name": "amenokaku-code-instruct" - }, - { - "url": "https://huggingface.co/datasets/kanhatakeyama/AutoMultiTurnByMixtral8x22b", - "downloads": 117, - "description": "自動生成のマルチターンデータセットオープンなデータソースから､MaziyarPanahi/Mixtral-8x22B-Instruct-v0.1-GGUFを使ってQ&Aを自動生成したものです｡関連コード一部の計算には東京工業大学のスーパーコンピュータTSUBAME4.0を利用しました｡データソースはじめの質問(q1)を､種々のデータソースから収集しました｡その後のやりとりはすべて､Mixtralが生成しました｡質問文については､元データのライセンスに準拠します｡oasst2-33k-jaapache 2.0databricks-dolly-15k-jacc-by-sa-3.0minnadeCC0cyberagent/chatbot-arena-ja-calm2-7b-chat-experimentalcc-by-4.0", + "description": "Japanese Wikipedia Human Retrieval datasetThis is a Japanese question answereing dataset with retrieval on Wikipedia articlesby trained human workers.", "source": "Hugging Face", - "score": 1.6259825903125625e-05, - "project_name": "AutoMultiTurnByMixtral8x22b" + "score": 1.8403492534017284e-05, + "project_name": "wikipedia-human-retrieval-ja" }, { - "url": "https://huggingface.co/datasets/Kendamarron/jimba-instuction-1k-beta", + "url": "https://huggingface.co/datasets/llm-book/jsnli", "downloads": 111, - "description": "cyberagent/calm2-7b-chatの出力を人手でチェック・修正することで作成した日本語Instructionデータセットです。", - "source": "Hugging Face", - "score": 1.542598867732431e-05, - "project_name": "jimba-instuction-1k-beta" - }, - { - "url": "https://huggingface.co/datasets/stockmark/ner-wikipedia-dataset", - "downloads": 110, - "description": "Wikipediaを用いた日本語の固有表現抽出データセットGitHub: https://github.com/stockmarkteam/ner-wikipedia-dataset/LICENSE: CC-BY-SA 3.0Developed by Stockmark Inc.", + "description": "JSNLI Version 1.1 のデータセットのうち、フィルタリング後の訓練セット (train_w_filtering)", "source": "Hugging Face", - "score": 1.5287015806357424e-05, - "project_name": "ner-wikipedia-dataset" + "score": 1.6608029847771696e-05, + "project_name": "jsnli" }, { - "url": "https://huggingface.co/datasets/range3/cc100-ja", + "url": "https://huggingface.co/datasets/HachiML/alpaca_jp_python", "downloads": 106, - "description": "range3/cc100-jaThis dataset consists of parquet files from the cc100 dataset with only the Japanese language extracted and sharded.", + "description": "alpaca_jp_pythonalpaca_jp_pythonは、Stanford Alpacaの手法mistralai/Mixtral-8x22B-Instruct-v0.1で作った合成データ(Synthetic data)です。", "source": "Hugging Face", - "score": 1.4731124322489883e-05, - "project_name": "cc100-ja" + "score": 1.5859920395169366e-05, + "project_name": "alpaca_jp_python" }, { "url": "https://huggingface.co/datasets/p1atdev/ja-stackoverflow", "downloads": 105, "description": "ja-stackoverflow日本語版 Stack Overflow の スタック・オーバーフロー のデータダンプ をもとにデータを加工し、質問文と回答文のペアになるように調整した QA データセット。", "source": "Hugging Face", - "score": 1.4592151451522996e-05, + "score": 1.5710298504648903e-05, "project_name": "ja-stackoverflow" }, + { + "url": "https://huggingface.co/datasets/matsuxr/JaGovFaqs-22k", + "downloads": 97, + "description": "このデータセットについてこのデータは、日本の官公庁のWebサイトに掲載されている「よくある質問」を手作業で抽出し、インストラクション用のデータセットとしたものです。", + "source": "Hugging Face", + "score": 1.4513323380485176e-05, + "project_name": "JaGovFaqs-22k" + }, { "url": "https://huggingface.co/datasets/Verah/JParaCrawl-Filtered-English-Japanese-Parallel-Corpus", - "downloads": 104, + "downloads": 90, "description": "IntroductionThis is a LLM-filtered set of the first 1M rows from ntt's JParaCrawl v3 large English-Japanese parallel corpus.", "source": "Hugging Face", - "score": 1.4453178580556112e-05, + "score": 1.3465970146841915e-05, "project_name": "JParaCrawl-Filtered-English-Japanese-Parallel-Corpus" }, { - "url": "https://huggingface.co/datasets/hatakeyama-llm-team/AutoGeneratedJapaneseQA-CC", - "downloads": 103, - "description": "自動生成Q&Aデータソースから､MaziyarPanahi/Mixtral-8x22B-Instruct-v0.1-GGUFを使ってQ&Aを自動生成したものです｡Common Crawlをもとに生成しています。 ", + "url": "https://huggingface.co/datasets/yongtae-jp/orca_dpo_pairs_ja", + "downloads": 89, + "description": "About this datasetThis dataset is a machine translation of the Intel/orca_dpo_pairs dataset with Palm 2 (prompt for translation is pasted below).", "source": "Hugging Face", - "score": 1.4314205709589225e-05, - "project_name": "AutoGeneratedJapaneseQA-CC" + "score": 1.331634825632145e-05, + "project_name": "orca_dpo_pairs_ja" }, { - "url": "https://huggingface.co/datasets/llm-book/jsnli", - "downloads": 101, - "description": "JSNLI Version 1.1 のデータセットのうち、フィルタリング後の訓練セット (train_w_filtering)", + "url": "https://huggingface.co/datasets/oshizo/ASRClustering-ja", + "downloads": 89, + "description": "埋め込みモデルの学習、評価のためのクラスタリングデータセットです。", "source": "Hugging Face", - "score": 1.4036259967655454e-05, - "project_name": "jsnli" + "score": 1.331634825632145e-05, + "project_name": "ASRClustering-ja" }, { - "url": "https://huggingface.co/datasets/fujiki/japanese_hh-rlhf-49k", - "downloads": 98, - "description": "This is a little bit different version of kunishou/hh-rlhf-49k-ja without ng_translation == 1 examples.", + "url": "https://huggingface.co/datasets/ryo0634/bsd_ja_en", + "downloads": 88, + "description": "The dataset was constructed in 3 steps:selecting business scenes,writing monolingual conversation scenarios according to the selected scenes, andtranslating the scenarios into the other language.", "source": "Hugging Face", - "score": 1.3619341354754797e-05, - "project_name": "japanese_hh-rlhf-49k" + "score": 1.3166726365800985e-05, + "project_name": "bsd_ja_en" }, { - "url": "https://huggingface.co/datasets/Mitsua/wikidata-parallel-descriptions-en-ja", - "downloads": 96, - "description": "Wikidata parallel descriptions en-jaParallel corpus for machine translation generated from wikidata dump (2024-05-06).", + "url": "https://huggingface.co/datasets/oshizo/LawClustering-ja", + "downloads": 87, + "description": "埋め込みモデルの学習、評価のためのクラスタリングデータセットです。", "source": "Hugging Face", - "score": 1.3341395612821026e-05, - "project_name": "wikidata-parallel-descriptions-en-ja" + "score": 1.301710447528052e-05, + "project_name": "LawClustering-ja" }, { - "url": "https://huggingface.co/datasets/ryota39/Aya_ja", - "downloads": 95, - "description": "Aya_jaこのデータセットはCohereForAI/aya_datasetの日本語インストラクションデータのみを抽出したデータセットです。", + "url": "https://huggingface.co/datasets/kunishou/amenokaku-code-instruct", + "downloads": 83, + "description": "Amenokaku-Code-InstructUpdate:2023/12/27データセットに JaxTon , プロになるJava のコードデータ 180 レコードを追加しました。", "source": "Hugging Face", - "score": 1.320242274185414e-05, - "project_name": "Aya_ja" + "score": 1.2418616913198655e-05, + "project_name": "amenokaku-code-instruct" }, { - "url": "https://huggingface.co/datasets/matsuxr/JaGovFaqs-22k", - "downloads": 87, - "description": "このデータセットについてこのデータは、日本の官公庁のWebサイトに掲載されている「よくある質問」���手作業で抽出し、インストラクション用のデータセットとしたものです。", + "url": "https://huggingface.co/datasets/Mitsua/wikidata-parallel-descriptions-en-ja", + "downloads": 77, + "description": "Wikidata parallel descriptions en-jaParallel corpus for machine translation generated from wikidata dump (2024-05-06).", "source": "Hugging Face", - "score": 1.2090639774119054e-05, - "project_name": "JaGovFaqs-22k" + "score": 1.1520885570075861e-05, + "project_name": "wikidata-parallel-descriptions-en-ja" }, { - "url": "https://huggingface.co/datasets/tohoku-nlp/abc-multiple-choice", - "downloads": 87, - "description": "abc-multiple-choice Datasetabc-multiple-choice は、競技クイズの大会「abc」で使用された4択問題を元に作成された、多肢選択式の質問応答データセットです。", + "url": "https://huggingface.co/datasets/oshizo/HSClustering-ja", + "downloads": 77, + "description": "埋め込みモデルの学習、評価のためのクラスタリングデータセットです。", "source": "Hugging Face", - "score": 1.2090639774119054e-05, - "project_name": "abc-multiple-choice" + "score": 1.1520885570075861e-05, + "project_name": "HSClustering-ja" }, { - "url": "https://huggingface.co/datasets/SakanaAI/JA-VG-VQA-500", - "downloads": 85, - "description": "JA-VG-VQA-500Dataset DescriptionJA-VG-VQA-500 is a 500-sample subset of Japanese Visual Genome VQA dataset.", + "url": "https://huggingface.co/datasets/tohoku-nlp/abc-multiple-choice", + "downloads": 76, + "description": "abc-multiple-choice Datasetabc-multiple-choice は、競技クイズの大会「abc」で使用された4択問題を元に作成された、多肢選択式の質問応答データセットです。", "source": "Hugging Face", - "score": 1.1812694032185283e-05, - "project_name": "JA-VG-VQA-500" + "score": 1.1371263679555395e-05, + "project_name": "abc-multiple-choice" }, { "url": "https://huggingface.co/datasets/llm-book/jawiki-paragraphs", - "downloads": 84, + "downloads": 73, "description": "GitHub リポジトリ singletongue/wikipedia-utils で公開されているデータセットを利用しています。", "source": "Hugging Face", - "score": 1.1673721161218397e-05, + "score": 1.0922398007993999e-05, "project_name": "jawiki-paragraphs" }, { - "url": "https://huggingface.co/datasets/yulanfmy/databricks-qa-ja", - "downloads": 78, - "description": "データセット概要手動で作成したDatabricksに関する質問と回答ペアの日本語データセットです。", + "url": "https://huggingface.co/datasets/community-datasets/covid_tweets_japanese", + "downloads": 70, + "description": "The annotation is by majority decision by 5 - 10 crowd workers.", "source": "Hugging Face", - "score": 1.0839883935417084e-05, - "project_name": "databricks-qa-ja" + "score": 1.0473532336432601e-05, + "project_name": "covid_tweets_japanese" }, { "url": "https://huggingface.co/datasets/kunishou/oasst2-chat-68k-ja", - "downloads": 76, + "downloads": 69, "description": "oasst2-135k-jaをチャット形式に変換したデータセットになります。", "source": "Hugging Face", - "score": 1.0561938193483311e-05, + "score": 1.0323910445912135e-05, "project_name": "oasst2-chat-68k-ja" }, { - "url": "https://huggingface.co/datasets/HachiML/Evol-Alpaca-gen3-500", - "downloads": 76, - "description": "Evol-Alpaca-gen3-500Evol-Alpaca-gen3-500は、Stanford Alpacaのseed tasksを日本語化Evol-Instructionの手法mistralai/Mixtral-8x22B-Instruct-v0.1で作った合成データ(Synthetic data)です。", + "url": "https://huggingface.co/datasets/kunishou/oasst1-chat-44k-ja", + "downloads": 66, + "description": "oasst1-89k-jaをチャット形式に変換したデータセットになります。", "source": "Hugging Face", - "score": 1.0561938193483311e-05, - "project_name": "Evol-Alpaca-gen3-500" + "score": 9.875044774350739e-06, + "project_name": "oasst1-chat-44k-ja" }, { - "url": "https://huggingface.co/datasets/llm-jp/oasst2-33k-ja", - "downloads": 69, - "description": "oasst2-33k-jaThis repository provides an instruction tuning dataset developed by LLM-jp, a collaborative project launched in Japan.", + "url": "https://huggingface.co/datasets/Kendamarron/jimba-instuction-1k-beta", + "downloads": 65, + "description": "cyberagent/calm2-7b-chatの出力を人手でチェック・修正することで作成した日本語Instructionデータセットです。", "source": "Hugging Face", - "score": 9.589128096715112e-06, - "project_name": "oasst2-33k-ja" + "score": 9.725422883830273e-06, + "project_name": "jimba-instuction-1k-beta" }, { - "url": "https://huggingface.co/datasets/kunishou/oasst1-chat-44k-ja", - "downloads": 66, - "description": "oasst1-89k-jaをチャット形式に変換したデータセットになります。", + "url": "https://huggingface.co/datasets/hotchpotch/JaCWIR", + "downloads": 62, + "description": "JaCWIR: Japanese Casual Web IR - 日本語情報検索評価のための小規模でカジュアルなWebタイトルと概要のデータセット近年、大規模言語モデル（LLM）の台頭により、一般的な日本語を用いた自然な検索クエリで質問するユースケースが増えています。", "source": "Hugging Face", - "score": 9.172209483814455e-06, - "project_name": "oasst1-chat-44k-ja" + "score": 9.276557212268875e-06, + "project_name": "JaCWIR" }, { "url": "https://huggingface.co/datasets/izumi-lab/llm-japanese-dataset-vanilla", "downloads": 60, "description": "llm-japanese-dataset-vanillaLLM構築用の日本語チャットデータセットizumi-lab/llm-japanese-dataset から，日英翻訳のデータセット等を抜いたものです．", "source": "Hugging Face", - "score": 8.33837225801314e-06, + "score": 8.977313431227943e-06, "project_name": "llm-japanese-dataset-vanilla" }, { - "url": "https://huggingface.co/datasets/svjack/pokemon-blip-captions-en-ja", - "downloads": 57, - "description": "Dataset used to train Pokémon text to image model, add a Japanese Column of Pokémon BLIP captionsBLIP generated captions for Pokémon images from Few Shot Pokémon dataset introduced by Towards Faster and Stabilized GAN Training for High-fidelity Few-shot Image Synthesis (FastGAN).", + "url": "https://huggingface.co/datasets/llm-jp/oasst2-33k-ja", + "downloads": 60, + "description": "oasst2-33k-jaThis repository provides an instruction tuning dataset developed by LLM-jp, a collaborative project launched in Japan.", "source": "Hugging Face", - "score": 7.921453645112483e-06, - "project_name": "pokemon-blip-captions-en-ja" + "score": 8.977313431227943e-06, + "project_name": "oasst2-33k-ja" }, { - "url": "https://huggingface.co/datasets/kubota/defamation-japanese-twitter", - "downloads": 53, - "description": "defamation_japanese_twitterTwitter日本語誹謗中傷検出データセットDataset SummarySNSにおける誹謗中傷検出のためのデータセットです．", + "url": "https://huggingface.co/datasets/oshizo/JMDNClustering-ja", + "downloads": 60, + "description": "埋め込みモデルの学習、評価のためのクラスタリングデータセットです。", "source": "Hugging Face", - "score": 7.365562161244941e-06, - "project_name": "defamation-japanese-twitter" + "score": 8.977313431227943e-06, + "project_name": "JMDNClustering-ja" }, { - "url": "https://huggingface.co/datasets/hotchpotch/JaCWIR", - "downloads": 48, - "description": "JaCWIR: Japanese Casual Web IR - 日本語情報検索評価のための小規模でカジュアルなWebタイトルと概要のデータセット近年、大規模言語モデル（LLM）の台頭により、一般的な日本語を用いた自然な検索クエリで質問するユースケースが増えています。", + "url": "https://huggingface.co/datasets/yulanfmy/databricks-qa-ja", + "downloads": 59, + "description": "データセット概要手動で作成したDatabricksに関する質問と回答ペアの日本語データセットです。", "source": "Hugging Face", - "score": 6.670697806410513e-06, - "project_name": "JaCWIR" + "score": 8.827691540707479e-06, + "project_name": "databricks-qa-ja" }, { - "url": "https://huggingface.co/datasets/llm-book/aio-passages-bpr-bert-base-japanese-v3", - "downloads": 48, - "description": "llm-book/aio-passages のデータセットに対して、llm-book/bert-base-japanese-v3-bpr-passage-encoder によるパッセージのバイナリベクトルが embeddings フィールドに追加されています。", + "url": "https://huggingface.co/datasets/kanhatakeyama/AutoMultiTurnByMixtral8x22b", + "downloads": 59, + "description": "自動生成のマルチターンデータセットオープンなデータソースから､MaziyarPanahi/Mixtral-8x22B-Instruct-v0.1-GGUFを使ってQ&Aを自動生成したものです｡関連コード一部の計算には東京工業大学のスーパーコンピュータTSUBAME4.0を利用しました｡データソースはじめの質問(q1)を､種々のデータソースから収集しました｡その後のやりとりはすべて､Mixtralが生成しました｡質問文については､元データのライセンスに準拠します｡oasst2-33k-jaapache 2.0databricks-dolly-15k-jacc-by-sa-3.0minnadeCC0cyberagent/chatbot-arena-ja-calm2-7b-chat-experimentalcc-by-4.0", "source": "Hugging Face", - "score": 6.670697806410513e-06, - "project_name": "aio-passages-bpr-bert-base-japanese-v3" + "score": 8.827691540707479e-06, + "project_name": "AutoMultiTurnByMixtral8x22b" }, { - "url": "https://huggingface.co/datasets/p1atdev/ichikara-instruction", - "downloads": 47, - "description": "ichikara-instruction (Non Commercial)LLMのための日本語インストラクションデータ 公開ページ公開ページより、本データに関して、言語処理学会第３０回年次大会において発表を行います。", + "url": "https://huggingface.co/datasets/fujiki/japanese_hh-rlhf-49k", + "downloads": 57, + "description": "This is a little bit different version of kunishou/hh-rlhf-49k-ja without ng_translation == 1 examples.", "source": "Hugging Face", - "score": 6.531724935443627e-06, - "project_name": "ichikara-instruction" + "score": 8.528447759666547e-06, + "project_name": "japanese_hh-rlhf-49k" }, { "url": "https://huggingface.co/datasets/llm-book/ner-wikinews-dataset", - "downloads": 46, + "downloads": 56, "description": "固有表現ラベルはllm-book/ner-wikipedia-datasetと同様のものを採用しており、全部で8種類 (人名、法人名、地名、製品名、政治的組織名、施設名、その他の組織名、イベント名)あります。", "source": "Hugging Face", - "score": 6.392752064476741e-06, + "score": 8.378825869146081e-06, "project_name": "ner-wikinews-dataset" }, + { + "url": "https://huggingface.co/datasets/svjack/pokemon-blip-captions-en-ja", + "downloads": 54, + "description": "Dataset used to train Pokémon text to image model, add a Japanese Column of Pokémon BLIP captionsBLIP generated captions for Pokémon images from Few Shot Pokémon dataset introduced by Towards Faster and Stabilized GAN Training for High-fidelity Few-shot Image Synthesis (FastGAN).", + "source": "Hugging Face", + "score": 8.079582088105149e-06, + "project_name": "pokemon-blip-captions-en-ja" + }, { "url": "https://huggingface.co/datasets/taishi-i/nagisa_stopwords", - "downloads": 44, + "downloads": 53, "description": "Japanese stopwords for nagisaThis is a stopword list of frequently used words in the Japanese language, created according to the tokenization rules of the Japanese text analysis library, nagisa.", "source": "Hugging Face", - "score": 6.11480632254297e-06, + "score": 7.929960197584683e-06, "project_name": "nagisa_stopwords" }, { - "url": "https://huggingface.co/datasets/NilanE/ParallelFiction-Ja_En-100k", - "downloads": 43, - "description": "Dataset details:Each entry in this dataset is a sentence-aligned Japanese web novel chapter and English fan translation.", + "url": "https://huggingface.co/datasets/llm-book/aio-passages", + "downloads": 51, + "description": "GitHub リポジトリ cl-tohoku/quiz-datasets で公開されているデータセットを利用しています。", "source": "Hugging Face", - "score": 5.975833451576084e-06, - "project_name": "ParallelFiction-Ja_En-100k" + "score": 7.630716416543753e-06, + "project_name": "aio-passages" }, { - "url": "https://huggingface.co/datasets/ryo0634/bsd_ja_en", - "downloads": 43, - "description": "The dataset was constructed in 3 steps:selecting business scenes,writing monolingual conversation scenarios according to the selected scenes, andtranslating the scenarios into the other language.", + "url": "https://huggingface.co/datasets/hatakeyama-llm-team/AutoGeneratedJapaneseQA-CC", + "downloads": 48, + "description": "自動生成Q&Aデータソースから､MaziyarPanahi/Mixtral-8x22B-Instruct-v0.1-GGUFを使ってQ&Aを自動生成したものです｡Common Crawlをもとに生成しています。 ", "source": "Hugging Face", - "score": 5.975833451576084e-06, - "project_name": "bsd_ja_en" + "score": 7.181850744982355e-06, + "project_name": "AutoGeneratedJapaneseQA-CC" + }, + { + "url": "https://huggingface.co/datasets/llm-book/aio-passages-bpr-bert-base-japanese-v3", + "downloads": 46, + "description": "llm-book/aio-passages のデータセットに対して、llm-book/bert-base-japanese-v3-bpr-passage-encoder によるパッセージのバイナリベクトルが embeddings フィールドに追加されています。", + "source": "Hugging Face", + "score": 6.882606963941424e-06, + "project_name": "aio-passages-bpr-bert-base-japanese-v3" }, { "url": "https://huggingface.co/datasets/bclavie/mmarco-japanese-hard-negatives", - "downloads": 42, + "downloads": 43, "description": "[Under Construction]This is a repository containing all the queries from the Japanese part of the MMarco dataset, the multilingual version of the MSMarco dataset.", "source": "Hugging Face", - "score": 5.8368605806091985e-06, + "score": 6.433741292380027e-06, "project_name": "mmarco-japanese-hard-negatives" }, + { + "url": "https://huggingface.co/datasets/NilanE/ParallelFiction-Ja_En-100k", + "downloads": 43, + "description": "Dataset details:Each entry in this dataset is a sentence-aligned Japanese web novel chapter and English fan translation.", + "source": "Hugging Face", + "score": 6.433741292380027e-06, + "project_name": "ParallelFiction-Ja_En-100k" + }, { "url": "https://huggingface.co/datasets/kunishou/jp-effective-instructions", "downloads": 38, "description": "oasst1-89k-ja , databricks-dolly-15k-ja , hh-rlhf-49k-ja の中から JGLUE（ JcommonsenseQA , MARC-ja , JSQuAD ）の観点で高品質なデータセットに絞り込んだデータセットです。", "source": "Hugging Face", - "score": 5.280969096741656e-06, + "score": 5.685631839777698e-06, "project_name": "jp-effective-instructions" }, { - "url": "https://huggingface.co/datasets/sakusakumura/databricks-dolly-15k-ja-scored", - "downloads": 38, - "description": "For the English version, please click here.", + "url": "https://huggingface.co/datasets/SakanaAI/JA-VG-VQA-500", + "downloads": 37, + "description": "JA-VG-VQA-500Dataset DescriptionJA-VG-VQA-500 is a 500-sample subset of Japanese Visual Genome VQA dataset.", "source": "Hugging Face", - "score": 5.280969096741656e-06, - "project_name": "databricks-dolly-15k-ja-scored" + "score": 5.536009949257232e-06, + "project_name": "JA-VG-VQA-500" }, { - "url": "https://huggingface.co/datasets/llm-book/aio-passages", - "downloads": 38, - "description": "GitHub リポジトリ cl-tohoku/quiz-datasets で公開されているデータセットを利用しています。", + "url": "https://huggingface.co/datasets/sakusakumura/databricks-dolly-15k-ja-scored", + "downloads": 36, + "description": "For the English version, please click here.", "source": "Hugging Face", - "score": 5.280969096741656e-06, - "project_name": "aio-passages" + "score": 5.3863880587367665e-06, + "project_name": "databricks-dolly-15k-ja-scored" }, { "url": "https://huggingface.co/datasets/kunishou/HelpSteer-35k-ja", "downloads": 32, "description": "NVIDIA が公開している SteerLM 向けのトライアルデータセット HelpSteerを日本語に自動翻訳したデータセットになります。", "source": "Hugging Face", - "score": 4.447131870940342e-06, + "score": 4.7879004966549035e-06, "project_name": "HelpSteer-35k-ja" }, { - "url": "https://huggingface.co/datasets/hpprc/jawiki", + "url": "https://huggingface.co/datasets/HachiML/Evol-Alpaca-gen3-500", + "downloads": 32, + "description": "Evol-Alpaca-gen3-500Evol-Alpaca-gen3-500は、Stanford Alpacaのseed tasksを日本語化Evol-Instructionの手法mistralai/Mixtral-8x22B-Instruct-v0.1で作った合成データ(Synthetic data)です。", + "source": "Hugging Face", + "score": 4.7879004966549035e-06, + "project_name": "Evol-Alpaca-gen3-500" + }, + { + "url": "https://huggingface.co/datasets/SNOW-NLP/snow_simplified_japanese_corpus", "downloads": 31, - "description": "JaWikiWikipediaのHTML形式のダンプファイルから抽出したテキストデータセットです。", + "description": "The corpus has 50,000 manually simplified and aligned sentences.", "source": "Hugging Face", - "score": 4.3081589999734566e-06, - "project_name": "jawiki" + "score": 4.6382786061344375e-06, + "project_name": "snow_simplified_japanese_corpus" }, { - "url": "https://huggingface.co/datasets/community-datasets/covid_tweets_japanese", - "downloads": 30, - "description": "The annotation is by majority decision by 5 - 10 crowd workers.", + "url": "https://huggingface.co/datasets/p1atdev/ichikara-instruction", + "downloads": 29, + "description": "ichikara-instruction (Non Commercial)LLMのための日本語インストラクションデータ 公開ページ公開ページより、本データに関して、言語処理学会第３０回年次大会において発表を行います。", "source": "Hugging Face", - "score": 4.16918612900657e-06, - "project_name": "covid_tweets_japanese" + "score": 4.339034825093506e-06, + "project_name": "ichikara-instruction" }, { - "url": "https://huggingface.co/datasets/Nexdata/English-Japanese_Parallel_Corpus_Data", - "downloads": 30, - "description": "It covers multiple fields such as tourism, medical treatment, daily life, news, etc. ", + "url": "https://huggingface.co/datasets/Atsushi/fungi_indexed_mycological_papers_japanese", + "downloads": 29, + "description": "fungi_indexed_mycological_papers_japanese大菌輪「論文3行まとめ」データセット最終更新日：2024/6/3（R3-11757まで）LanguagesJapaneseThis dataset is available in Japanese only.", "source": "Hugging Face", - "score": 4.16918612900657e-06, - "project_name": "English-Japanese_Parallel_Corpus_Data" + "score": 4.339034825093506e-06, + "project_name": "fungi_indexed_mycological_papers_japanese" + }, + { + "url": "https://huggingface.co/datasets/y2lan/japan-law", + "downloads": 29, + "description": "Japanese LawsThis dataset comprises 8.75K law records retrieved from the official Japanese government website e-Gov. ", + "source": "Hugging Face", + "score": 4.339034825093506e-06, + "project_name": "japan-law" }, { "url": "https://huggingface.co/datasets/alfredplpl/anime-with-caption-cc0", "downloads": 29, "description": "Anime with caption CC-0 datasetこのデータセットはイラストに対する日本語キャプションを倫理的に学習しやすくするためのデータセットです。", "source": "Hugging Face", - "score": 4.030213258039685e-06, + "score": 4.339034825093506e-06, "project_name": "anime-with-caption-cc0" }, { - "url": "https://huggingface.co/datasets/GENIAC-Team-Ozaki/WikiHowNFQA-ja_cleaned", + "url": "https://huggingface.co/datasets/Nexdata/English-Japanese_Parallel_Corpus_Data", + "downloads": 29, + "description": "It covers multiple fields such as tourism, medical treatment, daily life, news, etc. ", + "source": "Hugging Face", + "score": 4.339034825093506e-06, + "project_name": "English-Japanese_Parallel_Corpus_Data" + }, + { + "url": "https://huggingface.co/datasets/kubota/defamation-japanese-twitter", "downloads": 27, + "description": "defamation_japanese_twitterTwitter日本語誹謗中傷検出データセットDataset SummarySNSにおける誹謗中傷検出のためのデータセットです．", + "source": "Hugging Face", + "score": 4.0397910440525745e-06, + "project_name": "defamation-japanese-twitter" + }, + { + "url": "https://huggingface.co/datasets/GENIAC-Team-Ozaki/WikiHowNFQA-ja_cleaned", + "downloads": 26, "description": "Lurunchik/WikiHowNFQAを日本語に翻訳し、人手でクリーニングしたデータセットです。", "source": "Hugging Face", - "score": 3.7522675161059134e-06, + "score": 3.890169153532109e-06, "project_name": "WikiHowNFQA-ja_cleaned" }, { - "url": "https://huggingface.co/datasets/turing-motors/Japanese-Heron-Bench", - "downloads": 27, - "description": "Japanese-Heron-BenchDataset DescriptionJapanese-Heron-Bench is a benchmark for evaluating Japanese VLMs (Vision-Language Models).", + "url": "https://huggingface.co/datasets/DataPilot/databricks-dolly-15k-Nyan-ja", + "downloads": 26, + "description": "このデータセットはkunishou氏が公開している\"databricks-dolly-15k\"を日本語訳したkunishou/databricks-dolly-15k-jaデータセットの語尾をArrowPro-7B-KUJIRAを用いて「にゃん！", "source": "Hugging Face", - "score": 3.7522675161059134e-06, - "project_name": "Japanese-Heron-Bench" + "score": 3.890169153532109e-06, + "project_name": "databricks-dolly-15k-Nyan-ja" }, { "url": "https://huggingface.co/datasets/kanhatakeyama/SyntheticText", "downloads": 25, "description": "以下のデータ源からランダムに抽出したテキストをもとに､phi3で再生成した文章です｡WikibooksWikipediaCosmopedia判例データコードこちら一部の計算には東京工業大学のスーパーコンピュータTSUBAME4.0を利用しました｡", "source": "Hugging Face", - "score": 3.474321774172142e-06, + "score": 3.7405472630116434e-06, "project_name": "SyntheticText" }, - { - "url": "https://huggingface.co/datasets/sudy-super/CoTangent", - "downloads": 24, - "description": "CoTangentは人手で作成された高品質でクリーンな100セットの日本語CoT用データセットです。", - "source": "Hugging Face", - "score": 3.3353489032052565e-06, - "project_name": "CoTangent" - }, { "url": "https://huggingface.co/datasets/ikeno-ada/Japanese-English_translation_of_contents_HScodes", "downloads": 24, "description": "日本郵便が提供する「国際郵便　内容品の日英・中英訳、HSコード類」（2024/05/09）のデータに基づいています。", "source": "Hugging Face", - "score": 3.3353489032052565e-06, + "score": 3.5909253724911774e-06, "project_name": "Japanese-English_translation_of_contents_HScodes" }, + { + "url": "https://huggingface.co/datasets/Nexdata/Japanese-English_Parallel_Corpus_Data", + "downloads": 23, + "description": "For more details, please refer to the link: https://www.nexdata.ai/datasets/153?", + "source": "Hugging Face", + "score": 3.441303481970712e-06, + "project_name": "Japanese-English_Parallel_Corpus_Data" + }, + { + "url": "https://huggingface.co/datasets/stockmark/ner-wikipedia-dataset", + "downloads": 23, + "description": "Wikipediaを用いた日本語の固有表現抽出データセットGitHub: https://github.com/stockmarkteam/ner-wikipedia-dataset/LICENSE: CC-BY-SA 3.0Developed by Stockmark Inc.", + "source": "Hugging Face", + "score": 3.441303481970712e-06, + "project_name": "ner-wikipedia-dataset" + }, { "url": "https://huggingface.co/datasets/turing-motors/LLaVA-Pretrain-JA", - "downloads": 24, + "downloads": 22, "description": "Dataset DetailsDataset Type:Japanese LLaVA Pretrain is a localized version of the original LLaVA Pretrain dataset.", "source": "Hugging Face", - "score": 3.3353489032052565e-06, + "score": 3.2916815914502463e-06, "project_name": "LLaVA-Pretrain-JA" }, + { + "url": "https://huggingface.co/datasets/hatakeyama-llm-team/rlhf-ja-chatbot-arena-phi-3-medium", + "downloads": 21, + "description": "cyberagent/chatbot-arena-ja-calm2-7b-chat-experimentalのchosenの項目をmicrosoft/Phi-3-medium-4k-instructにより生成したデータで置き換えたデータセットです。", + "source": "Hugging Face", + "score": 3.1420597009297803e-06, + "project_name": "rlhf-ja-chatbot-arena-phi-3-medium" + }, { "url": "https://huggingface.co/datasets/kanhatakeyama/AutoWikiQA", - "downloads": 23, + "downloads": 21, "description": "Wikipedia日本語版からのQ&Aの自動生成Mixtral 8x22bのGGUF(5bit)をベースに､Wikipedia日本語版の記事から､自動生成コード1自動生成コード2を使ってQ&Aを作成しました｡計算には東京工業大学のスーパーコンピュータTSUBAME4.0を利用しました｡注意回答にハルシネーション等が含まれている可能性があるので､フィルタリングをかける必要があるかもしれません｡", "source": "Hugging Face", - "score": 3.1963760322383706e-06, + "score": 3.1420597009297803e-06, "project_name": "AutoWikiQA" }, { - "url": "https://huggingface.co/datasets/Nexdata/Japanese-English_Parallel_Corpus_Data", - "downloads": 23, - "description": "For more details, please refer to the link: https://www.nexdata.ai/datasets/153?", + "url": "https://huggingface.co/datasets/OmniAICreator/Japanese-Roleplay-Dialogues", + "downloads": 20, + "description": "Japanese-Roleplay-DialoguesThis is a dialogue corpus collected from Japanese role-playing forum (commonly known as \"なりきりチャット(narikiri chat)\").", "source": "Hugging Face", - "score": 3.1963760322383706e-06, - "project_name": "Japanese-English_Parallel_Corpus_Data" + "score": 2.9924378104093148e-06, + "project_name": "Japanese-Roleplay-Dialogues" }, { - "url": "https://huggingface.co/datasets/Atsushi/fungi_indexed_mycological_papers_japanese", - "downloads": 22, - "description": "fungi_indexed_mycological_papers_japanese大菌輪「論文3行まとめ」データセット最終更新日：2024/6/3（R3-11757まで）LanguagesJapaneseThis dataset is available in Japanese only.", + "url": "https://huggingface.co/datasets/shi3z/OpenOrcaJapanese", + "downloads": 20, + "description": "OpenOrcaデータセットの日本語翻訳版ですhttps://huggingface.co/datasets/Open-Orca/OpenOrca現在翻訳作業が続行中で、OpenOrca全体の1/5程度の翻訳が終わった状態でひとまず公開します。", "source": "Hugging Face", - "score": 3.057403161271485e-06, - "project_name": "fungi_indexed_mycological_papers_japanese" + "score": 2.9924378104093148e-06, + "project_name": "OpenOrcaJapanese" }, { "url": "https://huggingface.co/datasets/oshizo/japanese-wikipedia-paragraphs", - "downloads": 22, + "downloads": 18, "description": "A slightly modified version of the parsing and chunking method for singletongue/wikipedia-utils.", "source": "Hugging Face", - "score": 3.057403161271485e-06, + "score": 2.6931940293683833e-06, "project_name": "japanese-wikipedia-paragraphs" }, { - "url": "https://huggingface.co/datasets/kunishou/oasst1-89k-ja", - "downloads": 19, - "description": "This dataset was created by automatically translating \"OpenAssistant/oasst1\" into Japanese.", + "url": "https://huggingface.co/datasets/turing-motors/Japanese-Heron-Bench", + "downloads": 18, + "description": "Japanese-Heron-BenchDataset DescriptionJapanese-Heron-Bench is a benchmark for evaluating Japanese VLMs (Vision-Language Models).", "source": "Hugging Face", - "score": 2.640484548370828e-06, - "project_name": "oasst1-89k-ja" + "score": 2.6931940293683833e-06, + "project_name": "Japanese-Heron-Bench" }, { - "url": "https://huggingface.co/datasets/SNOW-NLP/snow_simplified_japanese_corpus", - "downloads": 19, - "description": "The corpus has 50,000 manually simplified and aligned sentences.", + "url": "https://huggingface.co/datasets/augmxnt/shisa-pretrain-en-ja-v1", + "downloads": 18, + "description": "This pre-training dataset was created for shisa-base-7b-v1.It is primarily composed of a DSIR sampling of MADLAD-400 JA/EN tokens in a 90%/10% ratio.", "source": "Hugging Face", - "score": 2.640484548370828e-06, - "project_name": "snow_simplified_japanese_corpus" + "score": 2.6931940293683833e-06, + "project_name": "shisa-pretrain-en-ja-v1" }, { - "url": "https://huggingface.co/datasets/shi3z/ja_conv_wikipedia_orion14B_100K", - "downloads": 18, - "description": "AbstructThis is a multi-turn conversation dataset generated from the Japanese Wikipedia dataset using Orion14B-Chat.", + "url": "https://huggingface.co/datasets/Atsushi/fungi_diagnostic_chars_comparison_japanese", + "downloads": 16, + "description": "fungi_diagnostic_chars_comparison_japanese大菌輪「識別形質まとめ」データセット最終更新日：2024/6/3（R3-11757まで）LanguagesJapaneseThis dataset is available in Japanese only.", "source": "Hugging Face", - "score": 2.5015116774039424e-06, - "project_name": "ja_conv_wikipedia_orion14B_100K" + "score": 2.3939502483274517e-06, + "project_name": "fungi_diagnostic_chars_comparison_japanese" }, { - "url": "https://huggingface.co/datasets/augmxnt/shisa-pretrain-en-ja-v1", + "url": "https://huggingface.co/datasets/AhmedSSabir/Japanese-wiki-dump-sentence-dataset", + "downloads": 16, + "description": "Dataset5M (5121625) clean Japanese full sentence with the context.", + "source": "Hugging Face", + "score": 2.3939502483274517e-06, + "project_name": "Japanese-wiki-dump-sentence-dataset" + }, + { + "url": "https://huggingface.co/datasets/Atsushi/fungi_trait_circus_database", "downloads": 14, - "description": "This pre-training dataset was created for shisa-base-7b-v1.It is primarily composed of a DSIR sampling of MADLAD-400 JA/EN tokens in a 90%/10% ratio.", + "description": "fungi_trait_circus_database大菌輪「Trait Circus」データセット（統制形質）最終更新日：2023/12/29LanguagesJapanese and EnglishPlease do not use this dataset for academic purposes for the time being. ", "source": "Hugging Face", - "score": 1.9456201935363996e-06, - "project_name": "shisa-pretrain-en-ja-v1" + "score": 2.0947064672865202e-06, + "project_name": "fungi_trait_circus_database" }, { - "url": "https://huggingface.co/datasets/y2lan/japan-law", + "url": "https://huggingface.co/datasets/shi3z/ja_conv_wikipedia_orion14B_100K", "downloads": 13, - "description": "Japanese LawsThis dataset comprises 8.75K law records retrieved from the official Japanese government website e-Gov. ", + "description": "AbstructThis is a multi-turn conversation dataset generated from the Japanese Wikipedia dataset using Orion14B-Chat.", "source": "Hugging Face", - "score": 1.806647322569514e-06, - "project_name": "japan-law" + "score": 1.9450845767660547e-06, + "project_name": "ja_conv_wikipedia_orion14B_100K" }, { - "url": "https://huggingface.co/datasets/toshi456/llava-bench-in-the-wild-ja", - "downloads": 13, - "description": "This dataset is the data that corrected the translation errors and untranslated data of the Japanese data in MBZUAI/multilingual-llava-bench-in-the-wild.", + "url": "https://huggingface.co/datasets/Nexdata/multi_language", + "downloads": 12, + "description": "SummaryThe dataset contains 25,000 hours of multi-language reading speech data.", "source": "Hugging Face", - "score": 1.806647322569514e-06, - "project_name": "llava-bench-in-the-wild-ja" + "score": 1.7954626862455887e-06, + "project_name": "multi_language" }, { "url": "https://huggingface.co/datasets/kanhatakeyama/SyntheticTextWikiTranslate", "downloads": 12, "description": "以下のデータ源からランダムに抽出した日本語のテキストをPhi-3で再生成し､更に自動英訳したコーパスです｡WikibooksWikipediaコードこちら一部の計算には東京工業大学のスーパーコンピュータTSUBAME4.0を利用しました｡", "source": "Hugging Face", - "score": 1.6676744516026283e-06, + "score": 1.7954626862455887e-06, "project_name": "SyntheticTextWikiTranslate" } ] \ No newline at end of file