Upstar ranks the fastest-rising GitHub repos every hour, across 69 topics. Browse topics

#local-llm

Local LLMs on GitHub, October 19–25, 2026

In Local LLMs, 78 repos ranked the week of October 19 and gained 9,046 stars between them. Generated-Computers/Locals took #1, gaining 320 stars. The week brought 42 repos new to the topic's ranking, among them Magnvs7/Promethevs at #30.

The ranking as the week’s last hourly run left it, on October 26 at 00:00 UTC. It does not change.

repos ranked
78
stars gained
9,046
ranked in #local-llm for the first time
42

The top 10

Ordered by stars gained over the seven days relative to each repo’s size, which is why a repo can rank above one that gained more.

  1. 1
    Generated-Computers/Locals Run many AI agents at once on your Mac, up to 3x faster. First week in #local-llm
    stars that week +320
  2. 2
    StartLuxLabs/StartLux-Decision Open decision models from 0.8B to 27B: a calibrated probability for every option in one pass, and a short thought only when unsure. Engine for vLLM, llama.cpp and MLX.
    stars that week +838
  3. 3
    JMS1717/ming-voice Low-latency local speech-to-speech voice assistant with Ming-3.0-audio-tiny on AMD and Nvidia GPUs. Easy setup on Windows and Linux First week in #local-llm
    stars that week +285
  4. 4
    mw00/project-maya Run GLM-5.3 (744B), GLM-5.3-Flash (321B), DeepSeek-V4.1-Flash (748B) and MiniMax-M3.1-Flash (428B) on your own GPUs - one engine, one dashboard, OpenAI- and Anthropic-compatible API.
    stars that week +820
  5. 5
    Indras-Mirror/Xolotl Xolotl: run DeepSeek-V4.1-Flash, all 552B, as a local coding agent on one gaming PC (RTX 4090 + 110 GB RAM): ~33 tok/s on agent tasks, ~41 in fast mode, 512K context. Plus GLM-5.3-Flash, Qwen3.8 and LFM3-Omni. MoE experts tiered across VRAM, pinned RAM and NVMe, with Xolotl Direct NVMe->VRAM… First week in #local-llm
    stars that week +230
  6. 6
    simplexai-labs/PolyStrata Run large AI models on your own PC: Qwen3.8-Flash-Next on Strata's engine, and GLM 5.3 Flash, DeepSeek V4 Flash, LFM2.5, Qwen3.6, Ornith 1.5, Gemma 4, Tiel Coder and MiniCPM-V on a second engine written for PolyStrata. One-click install for Windows and Linux, OpenAI/Anthropic API on localhost.
    stars that week +258
  7. 8
    cobanov/PS5LM llama.cpp on a jailbroken PS5: run any GGUF model natively on the console, starting with Qwen 3.8
    stars that week +150
  8. 9
    ReDChicagOTypewriteR/ShotAI 面向可信局域网的本地 AI 工作台:基于 Ollama 提供对话、视觉理解、文档问答、知识库检索和 Windows 图片创作。
    stars that week +60
  9. 10
    ahtx/kindi Kindi 0.6.1: fast local LLM inference on low-cost GPUs and CPUs. DeepSeek-V4.1-Flash (552B) at 18.7 tok/s on one RTX 3060 12 GB with a dual-Xeon DDR4 server. Ternary Bonsai 2B-27B, AH-Nanbeige4.3-33B, Nemotron 3.5 Lightning 30B-A3B. NVIDIA from the Quadro K420 to the RTX 50 series, AMD Radeon… First week in #local-llm
    stars that week +60

12 repos of this ranking were removed since and are not shown: deleted from GitHub or taken out by Upstar’s review. Their ranks stay empty.

Ranks 11 to 30

  1. 11Mafdet-AI/mafdet +54 stars that week
  2. 12justxor/deepseek-local-guide-ru +48 stars that week
  3. 13maximpri/mlxtop +43 stars that week
  4. 14CSUNX233/Sparkle-Qwen27b-Kvmem-ninfer +43 stars that week
  5. 15nicorobinreal/llmspeed +45 stars that week
  6. 18Soulmate-Halo/heterogeneous-gpu-pd-lab +99 stars that week
  7. 19abhi245y/PS5LM-Plus +67 stars that week
  8. 20syv-ai/HyperQwen +152 stars that week
  9. 21MakazhanAlpamys/Soup +304 stars that week
  10. 22MAZHARMIK/python-ai-agent-ollama +30 stars that week
  11. 23StayLameBro/backburner +86 stars that week
  12. 24ddalcu/mlx-serve +118 stars that week
  13. 25ollama/ollama +1.1k stars that week
  14. 26peonist-ai/halogen-flash-server +81 stars that week
  15. 27hardenedlinux/SonicBoom +25 stars that week
  16. 28llamastash/llamastash +36 stars that week
  17. 30Magnvs7/Promethevs +24 stars that week
Show the other 40 repos
  1. 32magnitudedev/magnitude +182 stars that week
  2. 33shawnshekari/strixite +21 stars that week
  3. 35MagentaHokageType/ollama-windows-model-library-manager +21 stars that week
  4. 36ChaserMeadowlark/local-ai-api-endpoint-manager-windows +20 stars that week
  5. 38cactus-compute/cactus +223 stars that week
  6. 40AlexsJones/llmfit +306 stars that week
  7. 41nobodywho-ooo/nobodywho +62 stars that week
  8. 42AtomicBot-ai/atomic-agent +90 stars that week
  9. 43zhongkaifu/TensorSharp +34 stars that week
  10. 44ollaya-dev/ollaya +49 stars that week
  11. 45feder-cr/jev +39 stars that week
  12. 46Osmantic/ODS +114 stars that week
  13. 47drumih/turbo-fieldfare +109 stars that week
  14. 49XHToken/Spark-X2.6 +33 stars that week
  15. 51Mintplex-Labs/anything-llm +292 stars that week
  16. 52prabanta-dev/janas +23 stars that week
  17. 53jundot/omlx +157 stars that week
  18. 54AtomicBot-ai/Atomic-Chat +41 stars that week
  19. 55ml-explore/mlx-lm +82 stars that week
  20. 56mostlygeek/llama-swap +72 stars that week
  21. 57PrismML-Eng/Bonsai-demo +45 stars that week
  22. 59ggml-org/Llama-macOS +30 stars that week
  23. 60isair/jarvis +32 stars that week
  24. 61youssofal/MTPLX +35 stars that week
  25. 62vllm-project/vllm-metal +26 stars that week
  26. 63fikrikarim/parlor +27 stars that week
  27. 64andrew-waters/orchard +24 stars that week
  28. 65ikawrakow/ik_llama.cpp +33 stars that week
  29. 66raullenchai/Rapid-MLX +36 stars that week
  30. 67janhq/jan +114 stars that week
  31. 69chatboxai/chatbox +107 stars that week
  32. 70off-grid-ai/OGAM +29 stars that week
  33. 71mozilla-ai/llamafile +78 stars that week
  34. 72cactus-compute/engine +36 stars that week
  35. 73Luce-Org/lucebox +23 stars that week
  36. 74Andyyyy64/whichllm +30 stars that week
  37. 75osaurus-ai/osaurus +30 stars that week
  38. 76Neywa-Labs/mlx-vlm +21 stars that week
  39. 77n4ze3m/page-assist +24 stars that week
  40. 78oobabooga/textgen +22 stars that week