andrijdavid commited on
Commit
b252247
1 Parent(s): 31681d3

Upload folder using huggingface_hub

Browse files
.gitattributes CHANGED
@@ -33,3 +33,17 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ Q2_K/Q2_K-00001-of-00001.gguf filter=lfs diff=lfs merge=lfs -text
37
+ Q3_K_L/Q3_K_L-00001-of-00001.gguf filter=lfs diff=lfs merge=lfs -text
38
+ Q3_K_M/Q3_K_M-00001-of-00001.gguf filter=lfs diff=lfs merge=lfs -text
39
+ Q3_K_S/Q3_K_S-00001-of-00001.gguf filter=lfs diff=lfs merge=lfs -text
40
+ Q4_0/Q4_0-00001-of-00001.gguf filter=lfs diff=lfs merge=lfs -text
41
+ Q4_1/Q4_1-00001-of-00001.gguf filter=lfs diff=lfs merge=lfs -text
42
+ Q4_K_M/Q4_K_M-00001-of-00001.gguf filter=lfs diff=lfs merge=lfs -text
43
+ Q4_K_S/Q4_K_S-00001-of-00001.gguf filter=lfs diff=lfs merge=lfs -text
44
+ Q5_0/Q5_0-00001-of-00001.gguf filter=lfs diff=lfs merge=lfs -text
45
+ Q5_1/Q5_1-00001-of-00001.gguf filter=lfs diff=lfs merge=lfs -text
46
+ Q5_K_M/Q5_K_M-00001-of-00001.gguf filter=lfs diff=lfs merge=lfs -text
47
+ Q5_K_S/Q5_K_S-00001-of-00001.gguf filter=lfs diff=lfs merge=lfs -text
48
+ Q6_K/Q6_K-00001-of-00001.gguf filter=lfs diff=lfs merge=lfs -text
49
+ Q8_0/Q8_0-00001-of-00001.gguf filter=lfs diff=lfs merge=lfs -text
Q2_K/Q2_K-00001-of-00001.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6e2434553508334864d5bc93921ab319c844bde079ee06bc70dac251a896a861
3
+ size 3179130976
Q3_K_L/Q3_K_L-00001-of-00001.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3eb89405ea57e5fa0a692827b2e8e0959c932d31155849617303dc34a03a2850
3
+ size 4321955936
Q3_K_M/Q3_K_M-00001-of-00001.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d562ea1ae02f9c39fd3e1d4a03e9647639b20133536e43064037f2bf27981168
3
+ size 4018917472
Q3_K_S/Q3_K_S-00001-of-00001.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e3bc564bfa53190a1a36d327b8a5231215617c78f78331269f990b04febc59e
3
+ size 3664498784
Q4_0/Q4_0-00001-of-00001.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1f2579d8b9dc7a61f01673ef9ba47a143c55f9f16f8d31b025bdebe41b3ef610
3
+ size 4661211232
Q4_1/Q4_1-00001-of-00001.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3283d38abf38d95aa5b98d77f2b71b355dd95019a9218961cd1bd402565ae91c
3
+ size 5130252384
Q4_K_M/Q4_K_M-00001-of-00001.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b43cb39cfbe5928495e33e8f278fecd1385e5c7198810102c219cf5ae59e54aa
3
+ size 4920733792
Q4_K_S/Q4_K_S-00001-of-00001.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:da2c89c3dae6a09ff9c4ca999ac09cd070dc3501dffac32121cce2ca3d336d67
3
+ size 4692668512
Q5_0/Q5_0-00001-of-00001.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:963ddbe75aebc6da7fc41c687b953fcc0884cf3ffc3a26941a8b047e1fbf4e98
3
+ size 5599293536
Q5_1/Q5_1-00001-of-00001.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1aaa1c6dda86ffb6584081a4ba8239102758584dc7d4ba2438893a785dcfe55e
3
+ size 6068334688
Q5_K_M/Q5_K_M-00001-of-00001.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1bf5a8e2785be9b4d6efdd1c04581d4bb745ef6beee9722ac0a655db0f685501
3
+ size 5732986976
Q5_K_S/Q5_K_S-00001-of-00001.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d79187d4ffba5dc431ec81be3dd053d26234fcde3c832ba0728d7c498758e66a
3
+ size 5599293536
Q6_K/Q6_K-00001-of-00001.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:35a0009820f954c255a23a4ed7e0efdcfa92f7db444582166174305cda7fc915
3
+ size 6596005984
Q8_0/Q8_0-00001-of-00001.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c1f8c6f7bc2450822410520f163c2b833d4cf1c816e136b80cfa9d33a7a8c5a9
3
+ size 8540770400
README.md CHANGED
@@ -215,6 +215,202 @@ Here are guides on using llama-cpp-python and ctransformers with LangChain:
215
 
216
 
217
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
218
  # OpenBuddy - Open Multilingual Chatbot
219
 
220
 
@@ -282,3 +478,5 @@ OpenBuddy按“原样”提供,不附带任何种类的明示或暗示的保
282
  使用OpenBuddy即表示您同意这些条款和条件,并承认您了解其使用可能带来的潜在风险。您还同意赔偿并使作者、贡献者和版权所有者免受因您使用OpenBuddy而产生的任何索赔、损害赔偿或责任的影响。
283
 
284
  <!-- original-model-card end -->
 
 
 
215
 
216
 
217
 
218
+ # openbuddy-llama3-8b-v21.1-8k-GGUF
219
+ - Original model: [openbuddy-llama3-8b-v21.1-8k](https://huggingface.co/OpenBuddy/openbuddy-llama3-8b-v21.1-8k)
220
+
221
+ <!-- description start -->
222
+ ## Description
223
+
224
+ This repo contains GGUF format model files for [openbuddy-llama3-8b-v21.1-8k](https://huggingface.co/OpenBuddy/openbuddy-llama3-8b-v21.1-8k).
225
+
226
+ <!-- description end -->
227
+ <!-- README_GGUF.md-about-gguf start -->
228
+ ### About GGUF
229
+ GGUF is a new format introduced by the llama.cpp team on August 21st 2023. It is a replacement for GGML, which is no longer supported by llama.cpp.
230
+ Here is an incomplete list of clients and libraries that are known to support GGUF:
231
+ * [llama.cpp](https://github.com/ggerganov/llama.cpp). This is the source project for GGUF, providing both a Command Line Interface (CLI) and a server option.
232
+ * [text-generation-webui](https://github.com/oobabooga/text-generation-webui), Known as the most widely used web UI, this project boasts numerous features and powerful extensions, and supports GPU acceleration.
233
+ * [Ollama](https://github.com/jmorganca/ollama) Ollama is a lightweight and extensible framework designed for building and running language models locally. It features a simple API for creating, managing, and executing models, along with a library of pre-built models for use in various applications​
234
+ * [KoboldCpp](https://github.com/LostRuins/koboldcpp), A comprehensive web UI offering GPU acceleration across all platforms and architectures, particularly renowned for storytelling.
235
+ * [GPT4All](https://gpt4all.io), This is a free and open source GUI that runs locally, supporting Windows, Linux, and macOS with full GPU acceleration.
236
+ * [LM Studio](https://lmstudio.ai/) An intuitive and powerful local GUI for Windows and macOS (Silicon), featuring GPU acceleration.
237
+ * [LoLLMS Web UI](https://github.com/ParisNeo/lollms-webui). A notable web UI with a variety of unique features, including a comprehensive model library for easy model selection.
238
+ * [Faraday.dev](https://faraday.dev/), An attractive, user-friendly character-based chat GUI for Windows and macOS (both Silicon and Intel), also offering GPU acceleration.
239
+ * [llama-cpp-python](https://github.com/abetlen/llama-cpp-python), A Python library equipped with GPU acceleration, LangChain support, and an OpenAI-compatible API server.
240
+ * [candle](https://github.com/huggingface/candle), A Rust-based ML framework focusing on performance, including GPU support, and designed for ease of use.
241
+ * [ctransformers](https://github.com/marella/ctransformers), A Python library featuring GPU acceleration, LangChain support, and an OpenAI-compatible AI server.
242
+ * [localGPT](https://github.com/PromtEngineer/localGPT) An open-source initiative enabling private conversations with documents.
243
+ <!-- README_GGUF.md-about-gguf end -->
244
+
245
+ <!-- compatibility_gguf start -->
246
+ ## Explanation of quantisation methods
247
+ <details>
248
+ <summary>Click to see details</summary>
249
+ The new methods available are:
250
+
251
+ * GGML_TYPE_Q2_K - "type-1" 2-bit quantization in super-blocks containing 16 blocks, each block having 16 weight. Block scales and mins are quantized with 4 bits. This ends up effectively using 2.5625 bits per weight (bpw)
252
+ * GGML_TYPE_Q3_K - "type-0" 3-bit quantization in super-blocks containing 16 blocks, each block having 16 weights. Scales are quantized with 6 bits. This end up using 3.4375 bpw.
253
+ * GGML_TYPE_Q4_K - "type-1" 4-bit quantization in super-blocks containing 8 blocks, each block having 32 weights. Scales and mins are quantized with 6 bits. This ends up using 4.5 bpw.
254
+ * GGML_TYPE_Q5_K - "type-1" 5-bit quantization. Same super-block structure as GGML_TYPE_Q4_K resulting in 5.5 bpw
255
+ * GGML_TYPE_Q6_K - "type-0" 6-bit quantization. Super-blocks with 16 blocks, each block having 16 weights. Scales are quantized with 8 bits. This ends up using 6.5625 bpw.
256
+ </details>
257
+ <!-- compatibility_gguf end -->
258
+
259
+ <!-- README_GGUF.md-how-to-download start -->
260
+ ## How to download GGUF files
261
+
262
+ **Note for manual downloaders:** You almost never want to clone the entire repo! Multiple different quantisation formats are provided, and most users only want to pick and download a single folder.
263
+
264
+ The following clients/libraries will automatically download models for you, providing a list of available models to choose from:
265
+
266
+ * LM Studio
267
+ * LoLLMS Web UI
268
+ * Faraday.dev
269
+
270
+ ### In `text-generation-webui`
271
+
272
+ Under Download Model, you can enter the model repo: LiteLLMs/openbuddy-llama3-8b-v21.1-8k-GGUF and below it, a specific filename to download, such as: Q4_0/Q4_0-00001-of-00009.gguf.
273
+
274
+ Then click Download.
275
+
276
+ ### On the command line, including multiple files at once
277
+
278
+ I recommend using the `huggingface-hub` Python library:
279
+
280
+ ```shell
281
+ pip3 install huggingface-hub
282
+ ```
283
+
284
+ Then you can download any individual model file to the current directory, at high speed, with a command like this:
285
+
286
+ ```shell
287
+ huggingface-cli download LiteLLMs/openbuddy-llama3-8b-v21.1-8k-GGUF Q4_0/Q4_0-00001-of-00009.gguf --local-dir . --local-dir-use-symlinks False
288
+ ```
289
+
290
+ <details>
291
+ <summary>More advanced huggingface-cli download usage (click to read)</summary>
292
+
293
+ You can also download multiple files at once with a pattern:
294
+
295
+ ```shell
296
+ huggingface-cli download LiteLLMs/openbuddy-llama3-8b-v21.1-8k-GGUF --local-dir . --local-dir-use-symlinks False --include='*Q4_K*gguf'
297
+ ```
298
+
299
+ For more documentation on downloading with `huggingface-cli`, please see: [HF -> Hub Python Library -> Download files -> Download from the CLI](https://huggingface.co/docs/huggingface_hub/guides/download#download-from-the-cli).
300
+
301
+ To accelerate downloads on fast connections (1Gbit/s or higher), install `hf_transfer`:
302
+
303
+ ```shell
304
+ pip3 install huggingface_hub[hf_transfer]
305
+ ```
306
+
307
+ And set environment variable `HF_HUB_ENABLE_HF_TRANSFER` to `1`:
308
+
309
+ ```shell
310
+ HF_HUB_ENABLE_HF_TRANSFER=1 huggingface-cli download LiteLLMs/openbuddy-llama3-8b-v21.1-8k-GGUF Q4_0/Q4_0-00001-of-00009.gguf --local-dir . --local-dir-use-symlinks False
311
+ ```
312
+
313
+ Windows Command Line users: You can set the environment variable by running `set HF_HUB_ENABLE_HF_TRANSFER=1` before the download command.
314
+ </details>
315
+ <!-- README_GGUF.md-how-to-download end -->
316
+ <!-- README_GGUF.md-how-to-run start -->
317
+ ## Example `llama.cpp` command
318
+
319
+ Make sure you are using `llama.cpp` from commit [d0cee0d](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221) or later.
320
+
321
+ ```shell
322
+ ./main -ngl 35 -m Q4_0/Q4_0-00001-of-00009.gguf --color -c 8192 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "<PROMPT>"
323
+ ```
324
+
325
+ Change `-ngl 32` to the number of layers to offload to GPU. Remove it if you don't have GPU acceleration.
326
+
327
+ Change `-c 8192` to the desired sequence length. For extended sequence models - eg 8K, 16K, 32K - the necessary RoPE scaling parameters are read from the GGUF file and set by llama.cpp automatically. Note that longer sequence lengths require much more resources, so you may need to reduce this value.
328
+
329
+ If you want to have a chat-style conversation, replace the `-p <PROMPT>` argument with `-i -ins`
330
+
331
+ For other parameters and how to use them, please refer to [the llama.cpp documentation](https://github.com/ggerganov/llama.cpp/blob/master/examples/main/README.md)
332
+
333
+ ## How to run in `text-generation-webui`
334
+
335
+ Further instructions can be found in the text-generation-webui documentation, here: [text-generation-webui/docs/04 ‐ Model Tab.md](https://github.com/oobabooga/text-generation-webui/blob/main/docs/04%20%E2%80%90%20Model%20Tab.md#llamacpp).
336
+
337
+ ## How to run from Python code
338
+
339
+ You can use GGUF models from Python using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) or [ctransformers](https://github.com/marella/ctransformers) libraries. Note that at the time of writing (Nov 27th 2023), ctransformers has not been updated for some time and is not compatible with some recent models. Therefore I recommend you use llama-cpp-python.
340
+
341
+ ### How to load this model in Python code, using llama-cpp-python
342
+
343
+ For full documentation, please see: [llama-cpp-python docs](https://abetlen.github.io/llama-cpp-python/).
344
+
345
+ #### First install the package
346
+
347
+ Run one of the following commands, according to your system:
348
+
349
+ ```shell
350
+ # Base ctransformers with no GPU acceleration
351
+ pip install llama-cpp-python
352
+ # With NVidia CUDA acceleration
353
+ CMAKE_ARGS="-DLLAMA_CUBLAS=on" pip install llama-cpp-python
354
+ # Or with OpenBLAS acceleration
355
+ CMAKE_ARGS="-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS" pip install llama-cpp-python
356
+ # Or with CLBLast acceleration
357
+ CMAKE_ARGS="-DLLAMA_CLBLAST=on" pip install llama-cpp-python
358
+ # Or with AMD ROCm GPU acceleration (Linux only)
359
+ CMAKE_ARGS="-DLLAMA_HIPBLAS=on" pip install llama-cpp-python
360
+ # Or with Metal GPU acceleration for macOS systems only
361
+ CMAKE_ARGS="-DLLAMA_METAL=on" pip install llama-cpp-python
362
+ # In windows, to set the variables CMAKE_ARGS in PowerShell, follow this format; eg for NVidia CUDA:
363
+ $env:CMAKE_ARGS = "-DLLAMA_OPENBLAS=on"
364
+ pip install llama-cpp-python
365
+ ```
366
+
367
+ #### Simple llama-cpp-python example code
368
+
369
+ ```python
370
+ from llama_cpp import Llama
371
+ # Set gpu_layers to the number of layers to offload to GPU. Set to 0 if no GPU acceleration is available on your system.
372
+ llm = Llama(
373
+ model_path="./Q4_0/Q4_0-00001-of-00009.gguf", # Download the model file first
374
+ n_ctx=32768, # The max sequence length to use - note that longer sequence lengths require much more resources
375
+ n_threads=8, # The number of CPU threads to use, tailor to your system and the resulting performance
376
+ n_gpu_layers=35 # The number of layers to offload to GPU, if you have GPU acceleration available
377
+ )
378
+ # Simple inference example
379
+ output = llm(
380
+ "<PROMPT>", # Prompt
381
+ max_tokens=512, # Generate up to 512 tokens
382
+ stop=["</s>"], # Example stop token - not necessarily correct for this specific model! Please check before using.
383
+ echo=True # Whether to echo the prompt
384
+ )
385
+ # Chat Completion API
386
+ llm = Llama(model_path="./Q4_0/Q4_0-00001-of-00009.gguf", chat_format="llama-2") # Set chat_format according to the model you are using
387
+ llm.create_chat_completion(
388
+ messages = [
389
+ {"role": "system", "content": "You are a story writing assistant."},
390
+ {
391
+ "role": "user",
392
+ "content": "Write a story about llamas."
393
+ }
394
+ ]
395
+ )
396
+ ```
397
+
398
+ ## How to use with LangChain
399
+
400
+ Here are guides on using llama-cpp-python and ctransformers with LangChain:
401
+
402
+ * [LangChain + llama-cpp-python](https://python.langchain.com/docs/integrations/llms/llamacpp)
403
+ * [LangChain + ctransformers](https://python.langchain.com/docs/integrations/providers/ctransformers)
404
+
405
+ <!-- README_GGUF.md-how-to-run end -->
406
+
407
+ <!-- footer end -->
408
+
409
+ <!-- original-model-card start -->
410
+ # Original model card: openbuddy-llama3-8b-v21.1-8k
411
+
412
+
413
+
414
  # OpenBuddy - Open Multilingual Chatbot
415
 
416
 
 
478
  使用OpenBuddy即表示您同意这些条款和条件,并承认您了解其使用可能带来的潜在风险。您还同意赔偿并使作者、贡献者和版权所有者免受因您使用OpenBuddy而产生的任何索赔、损害赔偿或责任的影响。
479
 
480
  <!-- original-model-card end -->
481
+
482
+ <!-- original-model-card end -->