GitHub repo leaderboard by stars, growth rate and activity.

• vllm-project

  #1vllm-project/vllm

  A high-throughput and memory-efficient inference and serving engine for LLMs

  Python91,383 starslast pushed 23 hours ago
• 1Panel-dev

  #21Panel-dev/MaxKB

  🔥 MaxKB is an open-source platform for building enterprise-grade agents. 强大易用的开源企业级智能体平台。

  Python22,750 starslast pushed 23 hours ago
• modelscope

  #3modelscope/ms-swift

  Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 600+ LLMs (Qwen3.6, DeepSeek-V4, GLM-5.1, InternLM3, Llama4, ...) and 300+ MLLMs (Qwen3-VL, Qwen3-Omni, InternVL3.5, Ovis2.5, GLM4.5v, Gemma4, Llava, Phi4, ...) (AAAI 2025).

  Python15,567 starslast pushed 2 days ago
• qualcomm

  #4qualcomm/GenieX

  Run frontier LLMs and VLMs locally on Qualcomm devices across NPU, GPU, and CPU with a few lines of code

  Rust8,358 starslast pushed Yesterday
• zilliztech

  #5zilliztech/deep-searcher

  Open Source Deep Research Alternative to Reason and Search on Private Data. Written in Python.

  Python8,259 starslast pushed 10 months ago
• johnbean393

  #6johnbean393/Sidekick

  A native macOS app that allows users to chat with a local LLM that can respond with information from files, folders and websites on your Mac without installing any other software. Powered by llama.cpp.

  Swift3,311 starslast pushed 4 months ago
• RunanywhereAI

  #7RunanywhereAI/wally

  Talk to your Mac, query your docs, no cloud required. On-device voice AI + RAG

  C++1,543 starslast pushed Yesterday
• syv-ai

  #8syv-ai/qwen38-27b-rtx3090

  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	vllm-project vllm-project/vllm A high-throughput and memory-efficient inference and serving engine for LLMs	Python	91,383		last pushed 23 hours ago
2	1Panel-dev 1Panel-dev/MaxKB 🔥 MaxKB is an open-source platform for building enterprise-grade agents. 强大易用的开源企业级智能体平台。	Python	22,750		last pushed 23 hours ago
3	modelscope modelscope/ms-swift Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 600+ LLMs (Qwen3.6, DeepSeek-V4, GLM-5.1, InternLM3, Llama4, ...) and 300+ MLLMs (Qwen3-VL, Qwen3-Omni, InternVL3.5, Ovis2.5, GLM4.5v, Gemma4, Llava, Phi4, ...) (AAAI 2025).	Python	15,567		last pushed 2 days ago
4	qualcomm qualcomm/GenieX Run frontier LLMs and VLMs locally on Qualcomm devices across NPU, GPU, and CPU with a few lines of code	Rust	8,358		last pushed Yesterday
5	zilliztech zilliztech/deep-searcher Open Source Deep Research Alternative to Reason and Search on Private Data. Written in Python.	Python	8,259		last pushed 10 months ago
6	johnbean393 johnbean393/Sidekick A native macOS app that allows users to chat with a local LLM that can respond with information from files, folders and websites on your Mac without installing any other software. Powered by llama.cpp.	Swift	3,311		last pushed 4 months ago
7	RunanywhereAI RunanywhereAI/wally Talk to your Mac, query your docs, no cloud required. On-device voice AI + RAG	C++	1,543		last pushed Yesterday
8	syv-ai syv-ai/qwen38-27b-rtx3090 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