AI Atlasby Verbito
Explore › Run & train AI

Run AI on your computer

Run open AI models privately on your own hardware. Every tool here is free and open source.

Free alternative to LM Studio, ChatGPT Plus, Claude Pro

180 tools

  1. Run AI models like Llama, Qwen and DeepSeek on your own computer
    ★ 186kstars
    ollama by ollama · Free alternative to LM Studio, ChatGPT Plus
    No codingDesktop appHealthy
  2. Run AI chat models fast on an ordinary computer
    ★ 133kstars
    llama.cpp by ggml-org · Free alternative to LM Studio
    Some setupCommand-line toolHealthy
  3. Serve AI models fast to many users at once
    ★ 96.3kstars
    vllm by vllm-project
    Some setupSelf-hosted appHealthy
  4. Chat with AI privately on your own computer, no internet needed
    ★ 77.3kstars
    gpt4all by nomic-ai · Free alternative to ChatGPT, LM Studio
    No codingDesktop appFair
  5. Run AI for text, images and voice on your own hardware
    ★ 50.2kstars
    LocalAI by mudler · Free alternative to OpenAI API
    Some setupSelf-hosted appHealthy
  6. Run very large AI models on a small graphics card
    ★ 41.7kstars
    airllm by lyogavin
    For developersCode libraryHealthy
  7. Find which AI models will run on your computer
    ★ 38.6kstars
    llmfit by AlexsJones
    Some setupCommand-line toolHealthy
  8. Serve AI models fast on GPUs
    ★ 38.4kstars
    sglang by sgl-project
    Some setupSelf-hosted appHealthy
  9. Manage a Linux server and run AI models from a web panel
    ★ 37.4kstars
    1Panel by 1Panel-dev
    Some setupSelf-hosted appHealthy
  10. Run AI models fast with the MAX engine and Mojo language
    ★ 30.1kstars
    modular by modular
    For developersCode libraryHealthy
  11. Run an AI model from a single downloadable file
    ★ 26.6kstars
    llamafile by mozilla-ai
    No codingDesktop appHealthy
  12. Run neural networks fast on phones
    ★ 24kstars
    ncnn by Tencent
    For developersCode libraryHealthy
  13. Run AI models fast on Apple Silicon Macs from the menu bar
    ★ 23.3kstars
    omlx by jundot
    No codingDesktop appHealthy
  14. Run AI models on phones, laptops and in web browsers
    ★ 23.3kstars
    mlc-llm by mlc-ai
    For developersCode libraryHealthy
  15. Run AI models fast on any device
    ★ 22.2kstars
    onnxruntime by microsoft
    For developersCode libraryHealthy
  16. Run AI models directly in a web browser
    ★ 19.4kstars
    web-llm by mlc-ai
    For developersCode libraryHealthy
  17. Run very large AI models on your desktop
    ★ 16.5kstars
    FreeToken by FlashML-org
    No codingDesktop appHealthy
  18. Run AI models on your phone
    ★ 16.3kstars
    MNN by alibaba
    No codingPhone appHealthy
  19. Set up a private local AI workspace in one go
    ★ 15.3kstars
    self-hosted-ai-starter-kit by n8n-io
    Some setupSelf-hosted appHealthy
  20. Run AI models fastest on NVIDIA GPUs
    ★ 15kstars
    TensorRT-LLM by NVIDIA
    For developersCode libraryHealthy
  21. Run AI models faster on NVIDIA graphics cards
    ★ 13.4kstars
    TensorRT by NVIDIA
    For developersCode libraryHealthy
  22. Run early LLaMA models on your computer (historic)
    ★ 12.9kstars
    dalai by cocktailpeanut
    Some setupCommand-line toolUse with care
  23. Run open models as an OpenAI-compatible API
    ★ 12.6kstars
    OpenLLM by bentoml
    Some setupSelf-hosted appHealthy
  24. Speed up AI model serving by reusing past work
    ★ 12.3kstars
    LMCache by LMCache
    For developersCode libraryHealthy
  25. Serve AI models in production (for engineers)
    ★ 11.1kstars
    server by triton-inference-server
    Some setupSelf-hosted appHealthy
  26. Run AI models fast on Intel hardware
    ★ 11.1kstars
    openvino by openvinotoolkit
    For developersCode libraryHealthy
  27. Serve language models in production
    ★ 10.9kstars
    text-generation-inference by huggingface
    Some setupSelf-hosted appUse with careArchived
  28. Run large models at home by sharing across computers
    ★ 10.7kstars
    petals by bigscience-workshop
    For developersCode libraryFair
  29. Use Ollama from Python code
    ★ 10.7kstars
    ollama-python by ollama
    For developersCode libraryHealthy
  30. Run AI locally in mobile and desktop apps
    ★ 10.3kstars
    runanywhere-sdks by RunanywhereAI
    For developersCode libraryHealthy
  31. Run large AI models faster on a consumer graphics card
    ★ 9.9kstars
    PowerInfer by Tiiny-AI
    For developersCode libraryFair
  32. Run and serve open AI models with one platform
    ★ 9.6kstars
    inference by xorbitsai
    Some setupSelf-hosted appHealthy
  33. Package AI models into standard containers
    ★ 9.5kstars
    cog by replicate
    Some setupCommand-line toolHealthy
  34. Run large models on a single GPU for batch jobs
    ★ 9.3kstars
    FlexLLMGen by FMInference
    For developersCode libraryUse with careArchived
  35. Serve AI models as APIs
    ★ 8.9kstars
    BentoML by bentoml
    For developersCode libraryHealthy
  36. Run and fine-tune AI models faster on Intel hardware
    ★ 8.8kstars
    ipex-llm by intel
    For developersCode libraryUse with careArchived
  37. Run AI models locally on Qualcomm phones and laptops
    ★ 8.5kstars
    GenieX by qualcomm
    For developersCode libraryHealthy
  38. Serve AI models at datacenter scale
    ★ 8.4kstars
    dynamo by ai-dynamo
    Some setupSelf-hosted appFair
  39. Install and run AI apps with one click
    ★ 8.4kstars
    pinokio by pinokiocomputer · Free alternative to Paid AI app bundles
    No codingDesktop appHealthy
  40. Speed up AI models
    ★ 8.3kstars
    optimate by nebuly-ai
    For developersCode libraryUse with careArchived

Showing 40 of 180. Search or open the full list for more.