Hugging Face
The leading platform for ML models, datasets, and demos — hub for the open-source AI community
tool
?
How to use this collection
Start with the problem you want to solve: running a model, building an application, or evaluating an answer. Browse the source material, compare its constraints with your workload, and use the discussion to share what you learned. Top combines staff picks and community activity; it is not a benchmark leaderboard.
Explore LLM tools, open-weight models, research papers and evaluations, with staff picks and community discussion.
llama.cpp
Efficient C/C++ LLM inference — run models locally on CPU and consumer GPUs with quantization
repo
Hugging Face Transformers
State-of-the-art ML library with thousands of pretrained models for text, vision, and audio
repo
Mistral 7B
Compact 7B model outperforming Llama 2 13B with grouped-query and sliding window attention
model
Llama 2: Open Foundation and Fine-Tuned Chat Models
Meta's open-source LLM family from 7B to 70B parameters, available for research and commercial use
model
llama.cpp b11368
spec : add probabilistic sampling for simple draft and MTP ( #27694 ) Make the drafter probabilistic and the target verify by rejection sampling Drop stale spec_draft_q before drafting Fallback to argmax sampling for grammar-constrained requests and adding flag for enabling…
release
llama.cpp b11366
ggml-quants : avoid invalid rounding in qkx3 scale search ( #29817 ) ggml-quants : avoid invalid rounding in qkx3 scale search The imatrix scale search can produce an infinite, NaN, or otherwise out-of-range value when the fitted minimum collapses to the maximum or makes the…
release
Language Models are Few-Shot Learners
The GPT-3 paper showing that scaling language models enables few-shot learning across tasks
paper
llama.cpp b11365
ggml-cpu : fix soft_max_back wrong output when dst aliases src1 ( #27096 ) ggml-cpu : fix soft_max_back wrong output when dst aliases src1 GGML_OP_SOFT_MAX_BACK is listed in ggml_op_can_inplace, so the graph allocator may assign dst to alias either src0 (dy) or src1 (y). The…
release
llama.cpp b11364
model: support nimble decision model ( #29844 ) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/52332773 macOS/iOS: macOS Apple Silicon (arm64) macOS Apple Silicon (arm64, KleidiAI enabled) DISABLED macOS Intel (x64) iOS XCFramework…
release
BERT: Pre-training of Deep Bidirectional Transformers
Bidirectional encoder representations that revolutionized NLP transfer learning
paper
llama.cpp b11362
metal : add tensor API flash attention kernel for F16 KV ( #29570 ) metal : add tensor API flash attention kernel for F16 KV cont : add tensor FA kernels for DK=DV=512 and DK=576, DV=512 cont : support attention sinks, ALiBi and logit softcap in the tensor FA kernel cont : add…
release
llama.cpp b11361
llama, server: add /v1/systemone API (models: laya, julia-1, lev, openjev, kev) ( #29818 ) init conversion convert: ok model loaded add server code improve conversion script support shared prompt prefix add docs, imorove UX a bit add vision support add openjev tiny model for…
release
Attention Is All You Need
The foundational paper introducing the Transformer architecture that powers all modern LLMs
paper
DeepSeek Chat
DeepSeek's chat with extended thinking mode — strong on math, code, and reasoning tasks
chat
llama.cpp b11355
vulkan: disable large matmul tile on Samsung GPUs with 32KB shared memory ( #28531 ) Assisted-by: Claude Opus Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/52313922 macOS/iOS: macOS Apple Silicon (arm64) macOS Apple Silicon (arm64,…
release
llama.cpp b11352
qwen4exp : optimize mask constructions ( #29824 ) qwen4exp : optimize mask constructions cont : apply the same change for GLM5-next Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/52303821 macOS/iOS: macOS Apple Silicon (arm64) macOS…
release
vllm v0.31.0rc5: [Misc] Add Transformers version upper bound in requirements (#59614)
Signed-off-by: Isotr0py Isotr0py@outlook.com Co-authored-by: Harry Mellor 19981378+hmellor@users.noreply.github.com (cherry picked from commit 58b3298 )
release
OpenAI Python SDK
Official Python client for the OpenAI API — GPT, DALL-E, Whisper, and embeddings
repo
LM Studio
Desktop app to run any GGUF model locally — one-click downloads, chat UI, and local server mode
chat
Open WebUI
Self-hosted Ollama and OpenAI-compatible chat UI — full featured, RAG, multimodal, and pipelines
chat
llama.cpp b11351
ggml : add alloc_buffer_n to buffer type interface ( #23671 ) ggml : add alloc_buffer_n to buffer type interface Add alloc_buffer_n method to ggml_backend_buffer_type_i interface, with a public API ggml_backend_buft_alloc_buffer_n. Default implementation in ggml-backend.cpp…
release
Hugging Face Spaces
Host and share ML demos with Gradio and Streamlit — free GPU access for interactive applications
demo
ollama v0.35.1
What's Changed feat: allow ten web searches per response by @ParthSareen in #18602 MLX: version bump by @dhiltgen in #18651 llama.cpp: version bump b11232 by @dhiltgen in #18652 create: support explicit model capabilities by @dhiltgen in #18708 Full Changelog :…
release