A Thunderbird add-on that brings ChatGPT, Gemini, Claude or Ollama into your mailbox to reply, proofread, translate, summarize and tag emails.
swuecho/chat
Self-hosted team chat web app in Go that connects to OpenAI, Azure, Claude, Gemini and Ollama models, with user management and rate limiting.
Open-source alternative to:ChatGPT Team
- Stars
- 574
- Forks
- 93
- Open issues
- 1
- Last commit
- 27 days ago
What is it?
This project is a multi-user chat web application for teams. It lets several users share one deployment, with the first registered account becoming the administrator. Conversations can be connected to OpenAI (including Azure), Claude, Gemini and locally hosted Ollama models, and each conversation starts with a system prompt that can be managed from a prompt library.
The app adds operational features aimed at shared or hosted use: a configurable rate limit (the default is 100 ChatGPT calls per 10 minutes), conversation snapshots that can be published as static share pages, and full-text search across saved conversations. Users can upload text files and multimedia files, subject to what the chosen model supports.
It also includes a built-in OpenAI-compatible gateway that issues virtual API keys for calling configured models, with streaming pass-through, per-key limits, request auditing and time-limited request/response samples. The mobile client is written in Flutter.
Key features
- 01Supports OpenAI (including Azure), Claude, Gemini and Ollama models.
- 02Multi-user management where the first registered user is the administrator.
- 03Configurable rate limiting per user or deployment.
- 04Shareable static pages generated from conversations, and full-text search over saved chats.
- 05Built-in OpenAI-compatible gateway with virtual API keys, request auditing and streaming.
Use cases
- Giving a small team a shared, self-hosted chat front end for several LLM providers.
- Hosting a chat service with per-user rate limits and account management.
- Running local Ollama models alongside cloud models in one interface.
- Exposing configured models to other tools through an OpenAI-compatible endpoint with revocable keys.
Quick start
- 1
Follow the deployment guide in docs/deployment_en.md
- 2
Read the local development guide in docs/dev_locally_en.md to run it from source
Things to know
- Needs access to an OpenAI, Claude, or Ollama model backend to be useful.
- Much of the README and documentation is written in Chinese.
In collections
Topics
FAQ
- What is chat?
- Self-hosted team chat web app in Go that connects to OpenAI, Azure, Claude, Gemini and Ollama models, with user management and rate limiting. This project is a multi-user chat web application for teams. It lets several users share one deployment, with the first registered account becoming the administrator. Conversations can be connected to OpenAI (including Azure), Claude, Gemini and locally hosted Ollama models, and each conversation starts with a system prompt that can be managed from a prompt library.
- Is chat free to use? What license does it have?
- chat is open source under the MIT license. Check the license terms before using it in commercial products.
- Is chat still maintained?
- Actively maintained. The last commit to chat was on Oct 11, 2026.
- What can I use chat for?
- Giving a small team a shared, self-hosted chat front end for several LLM providers. Hosting a chat service with per-user rate limits and account management. Running local Ollama models alongside cloud models in one interface. Exposing configured models to other tools through an OpenAI-compatible endpoint with revocable keys.
- How do I install chat?
- Follow the deployment guide in docs/deployment_en.md Read the local development guide in docs/dev_locally_en.md to run it from source
- Where can I get chat?
- The source code and installation instructions are on GitHub: https://github.com/swuecho/chat
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
A self-hosted AI gateway that exposes OpenAI, Claude and Gemini compatible APIs in front of many upstream model providers, with routing, keys, quotas and cost tracking.
Self-hosted gateway that exposes many LLM providers (OpenAI, Claude, Gemini, DeepSeek, Qwen, and more) behind one OpenAI-compatible API.
Self-hosted admin system for integrating many LLM providers (OpenAI, Claude, Gemini, DeepSeek, Qwen and others) behind one unified API.
A Python Telegram bot that bundles many AI models and media tools (ChatGPT, Claude, Gemini, ChatTTS, image and video generation) into one chat interface.
Windows desktop app that sends one prompt to ChatGPT, Gemini, Claude, Grok, DeepSeek and Copilot at the same time.