Host one chat interface in front of every model you use

Made by Danny Avila

A self-hosted chat application that puts a single interface in front of many model providers at once — OpenAI, Anthropic, Google, AWS, DeepSeek and others — so switching models does not mean switching product.

It supports agents you configure and share, MCP tools, skills, file search, code execution and artifacts. Recent releases add an API for managing agents programmatically, per-agent workspaces where an agent can inspect a file tree and run commands with time limits, and approval controls letting an administrator decide whether file writes and commands are allowed, asked about, or refused.

Can I use this?

Runs in Docker on your computer · needs an AI model key

You'll need A server or a computer you can run Docker on · Setup needed

Worth knowing You run and secure it, including the credentials for every provider it talks to and whatever conversation history it stores. Some of the most interesting features are labelled experimental by the project itself — attached workspaces are described as highly experimental — which is a fair warning about letting an agent run commands. The costs are still the providers’ costs; this changes the interface, not the bill. MIT licence. BuildTube has not run or verified this software.

Why it's here

Not trending this week

Last seen on the GitHub trending list on 20 September 2026. It stays on BuildTube so you can still find it.

45,332 people have starred it on GitHub.

Numbers checked on 6 October 2026; not refreshed since.

Behind it

See the code on GitHub

BuildTube has not run or verified this project. Everything above is written from what the creator published.

Made this?