GitHub Open Source Insight
Computed at Sep 11, 2026GitHub repo leaderboard by stars, growth rate and activity.
hiyouga
Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)
Python74,682 starslast pushed 2 days agoSYSTRAN
Faster Whisper transcription with CTranslate2
Python25,310 starslast pushed 10 months agoymcui
中文LLaMA&Alpaca大语言模型+本地CPU/GPU训练部署 (Chinese LLaMA & Alpaca LLMs)
Python18,942 starslast pushed 5 months agoRyanCodrai
A vector index built on TurboQuant, written in Rust with Python bindings
Rust16,734 starslast pushed 3 weeks agobitsandbytes-foundation
Accessible large language models via k-bit quantization for PyTorch.
Python8,466 starslast pushed 3 days agoFareedKhan-dev
A 2.78-trillion-parameter Kimi K3 running inference on a single CPU in 8.24 GB of RAM. Portable C99: no BLAS, no framework, no GPU.
C7,382 starslast pushed 2 weeks agoamitshekhariitbhu
Your Cheat Sheet for AI Engineering Interview – Questions and Answers.
Markdown3,037 starslast pushed Yesterdaysyv-ai
Qwen3.8-27B on a single RTX 3090 with vLLM: ~1,000 tok/s at 64 concurrent (int8 tensor-core GEMMs, fp16 DeltaNet state), ~114 tok/s single-user at default sampling / ~124 greedy (MTP drafts, own-output draft vocab, calibrated int4 lm_head, split-KV verify attention), 150k-262k context; patches, requant scripts, benchmarks
Python1,238 starslast pushed 3 days ago
| # | Repo | Language | Stars | 30-day trend | Last updated |
|---|---|---|---|---|---|
| 1 | hiyouga Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024) | Python | 74,682 | last pushed 2 days ago | |
| 2 | SYSTRAN Faster Whisper transcription with CTranslate2 | Python | 25,310 | last pushed 10 months ago | |
| 3 | ymcui 中文LLaMA&Alpaca大语言模型+本地CPU/GPU训练部署 (Chinese LLaMA & Alpaca LLMs) | Python | 18,942 | last pushed 5 months ago | |
| 4 | RyanCodrai A vector index built on TurboQuant, written in Rust with Python bindings | Rust | 16,734 | last pushed 3 weeks ago | |
| 5 | bitsandbytes-foundation Accessible large language models via k-bit quantization for PyTorch. | Python | 8,466 | last pushed 3 days ago | |
| 6 | FareedKhan-dev A 2.78-trillion-parameter Kimi K3 running inference on a single CPU in 8.24 GB of RAM. Portable C99: no BLAS, no framework, no GPU. | C | 7,382 | last pushed 2 weeks ago | |
| 7 | amitshekhariitbhu Your Cheat Sheet for AI Engineering Interview – Questions and Answers. | Markdown | 3,037 | last pushed Yesterday | |
| 8 | syv-ai Qwen3.8-27B on a single RTX 3090 with vLLM: ~1,000 tok/s at 64 concurrent (int8 tensor-core GEMMs, fp16 DeltaNet state), ~114 tok/s single-user at default sampling / ~124 greedy (MTP drafts, own-output draft vocab, calibrated int4 lm_head, split-KV verify attention), 150k-262k context; patches, requant scripts, benchmarks | Python | 1,238 | last pushed 3 days ago |
All · 11,474