open-webui  by open-webui

Self-hosted AI platform for local LLM deployment

Created 1 year ago
110,172 stars

Top 0.1% on SourcePulse

GitHubView on GitHub
Project Summary

Open WebUI provides a user-friendly, self-hosted AI platform for interacting with large language models (LLMs). It targets users who want a feature-rich, offline-capable interface for managing and conversing with various LLM runners, including Ollama and OpenAI-compatible APIs, with built-in RAG capabilities.

How It Works

The platform leverages a Docker-first deployment strategy, offering official images for various configurations including Ollama integration and CUDA acceleration. It supports OpenAI-compatible APIs, allowing flexible integration with services like LMStudio, GroqCloud, and Mistral. Key features include granular permissions, a responsive PWA for mobile, extensive Markdown/LaTeX support, and native Python function calling for custom tool integration.

Quick Start & Requirements

  • Install/Run:
    • Python: pip install open-webui then open-webui serve
    • Docker: docker run -d -p 3000:8080 --add-host=host.docker.internal:host-gateway -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:main (for Ollama on host)
    • Docker with GPU: docker run -d -p 3000:8080 --gpus all --add-host=host.docker.internal:host-gateway -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:cuda
  • Prerequisites: Python 3.11+ for pip install. Docker with NVIDIA Container Toolkit for GPU support.
  • Resources: Docker volume mounting (-v open-webui:/app/backend/data) is crucial for data persistence.
  • Docs: Open WebUI Documentation

Highlighted Details

  • Integrated Retrieval Augmented Generation (RAG) with document loading and web search capabilities (SearXNG, Brave, Bing, etc.).
  • Supports image generation via AUTOMATIC1111, ComfyUI, and DALL-E.
  • Extensible via a Pipelines Plugin Framework for custom logic and integrations.
  • Features hands-free voice/video calls and a model builder for creating custom agents.

Maintenance & Community

Licensing & Compatibility

  • License: BSD-3-Clause.
  • Permissive license suitable for commercial use and integration with closed-source applications.

Limitations & Caveats

  • The :dev Docker tag is for unstable, bleeding-edge features and may contain bugs.
  • Troubleshooting connection issues may require specific Docker network configurations (e.g., --network=host).
Health Check
Last Commit

1 day ago

Responsiveness

1 day

Pull Requests (30d)
264
Issues (30d)
315
Star History
3,435 stars in the last 30 days

Explore Similar Projects

Feedback? Help us improve.