llm-recipes

v1 = Build a ChatGPT like setup on local machine with Ollama + Open-WebUI + Mistral + Docker

• Requirements
  • CUDA installation - Needs NVIDIA Graphics Card
  • Docker Setup / Docker Desktop on Windows
  • VSCode Terminal
• Steps - (VSCode Terminal)
  • Clone Repo - https://github.com/slabstech/llm-recipes
    • git clone https://github.com/slabstech/llm-recipes.git
    • cd llm-recipes
  • From src/ollama directory, run the command
    • cd src/ollama
    • docker-compose.exe -f .\docker-compose.yml up -d
    • Verify if ollama backend is running at http://localhost:11434
  • Sign Up with any credentials at http://localhost:3000/auth/
  • To make UI accessible across local network
    • Windows
      • Enable Network Sharing
      • Create Firewall rule with command on Windows PowerShell
        • New-NetFirewallRule -DisplayName "UI for LLM" -Direction Inbound -Protocol TCP -LocalPort 3000 -Action Allow
      • Get IP address from CMD shell
        • ipconfig
        • Check for IPv4 Address for WIFI/LAN(Ethernet Adaptor)
      • Access from Mobile/Laptop Brower fron the local network
        • < YourIPAddress >:3000
• References
  • Download models and UI
    • https://www.youtube.com/watch?v=syR0fT0rkgY&t=314s
  • https://medium.com/@omargohan/using-ollama-in-your-ide-with-continue-e8cefeeee033
  • https://github.com/open-webui/open-webui
  • https://github.com/ollama
  • mistral.ai

TODO -

• Tutorial- LLM RECIPES

Provide link for Docker and Cuda installation.

Update steps for linux / GPU server

• open step
• how to build hardware and system build

Function calling and RAG with RLHF

This site is open source. Improve this page.