AI & ML interests
transformers, datasets, tokenizers, accelerate, peft, trl, diffusers, safetensors, huggingface_hub, evaluate, optimum, text-generation-inference, TGI, model hub, spaces, inference API, inference endpoints, autoTrain, model cards, dataset viewer, leaderboard, llama, llama-3, llama-3.1, llama-3.2, mistral, mixtral, moE, qwen, deepseek, deepseek-r1, phi, phi-4, gemma, gemma-2, command-r, falcon, mpt, gpt-neo, gpt-j, bloom, bloomz, starcoder, codellama, base model, instruct model, chat model, foundation model, pre-trained, fine-tuned, distilled, quantized, merged model, mergekit, fine-tuning, full fine-tuning, parameter-efficient fine-tuning, PEFT, LoRA, QLoRA, DoRA, AdaLoRA, prefix tuning, prompt tuning, p-tuning, IA3, instruction tuning, SFT, supervised fine-tuning, continued pretraining, domain adaptation, gradient checkpointing, gradient accumulation, mixed precision, fp16, bf16, 8-bit, 4-bit, bitsandbytes, deepspeed, ZeRO, FSDP, flash attention, xFormers, unsloth, RLHF, reinforcement learning from human feedback, DPO, direct preference optimization, IPO, KTO, PPO, reward model, preference dataset, alignment, safety tuning, constitutional AI, red teaming, vLLM, text-generation-inference, TGI, llama.cpp, ollama, tensorRT-LLM, onnx runtime, openvino, Triton, fastAPI, gradio, streamlit, chainlit, quantization, GPTQ, AWQ, GGUF, GGML, EXL2, smoothquant, KV cache, continuous batching, paged attention, speculative decoding, model parallelism, pipeline parallelism, tensor parallelism, perplexity, bleu, rouge, meteor, bertscore, mauve, helm, mmlu, hellaswag, arc, truthfulqa, humaneval, mbpp, gsm8k, bbh, alpaca eval, mt-bench, lmsys arena, openllm leaderboard, eval harness, lm-evaluation-harness, retrieval augmented generation, RAG, vector database, chroma, faiss, pinecone, weaviate, milvus, qdrant, embeddings, sentence-transformers, semantic search, reranker, cross-encoder, langchain, llamaindex, haystack, agent, tool use, function calling, ReAct, chain-of-thought, CoT, transformer, attention, self-attention, multi-head attention, causal LM, masked LM, encoder-decoder, decoder-only, rope, rotary positional embedding, ALiBi, sliding window attention, grouped query attention, GQA, multi-query attention, MQA, mixture of experts, MoE, sparse attention, state space model, Mamba, RWKV, retnet