Skip to content

Instantly share code, notes, and snippets.

@razhangwei

razhangwei/open_webui.md

Last active February 16, 2025 04:01
Show Gist options
  • Save razhangwei/c58bb0f162986d691d75cd33cf48dcc2 to your computer and use it in GitHub Desktop.
Save razhangwei/c58bb0f162986d691d75cd33cf48dcc2 to your computer and use it in GitHub Desktop.
Download ZIP
Open WebUI
Raw
open_webui.md

  • How to use Pipe?

    • Find one on the website and import it.
    • Configure the Valves (paramters), usually just the API key is needed.
    • The Pipe's corresponding model will show up in the model list.

  • Add models

    • Add openrouter api in connection and manually add models so that the default list didn't overwelme
    • Add gemini api in the conneciton, but the model list won't be loaded correctly, so it also needs to manually add models (open-webui/open-webui#8445)
  1. Install with docker compose
  2. Enable various pipelines
    • Admin Panel -> Connections -> OpenAI API: http://host.docker.internal:9099 and 0p3n-w3bu!
    • Find pipelines at https://openwebui.com/tools
    • OpenAI
    • Google/Gemini
    • Groq
  3. Enable RAG
    • choose external content extraction (tika)
  4. Enable STT and TTS
    • Set up Https (with Caddy)
    • Choose whipser model, e.g., WHISPER_MODEL=small
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment