What is CoAI used for?

CoAI is a multi-tenant AI platform that combines an LLM gateway for many providers with a chat interface, admin dashboard and credit or subscription billing. It is a good fit for running a shared AI chat service for a team or community, reselling model access with billing and consolidating several LLM API keys behind one gateway.

Yes. CoAI is open source under the Apache-2.0 licence. Its source code is on GitHub at coaidev/coai and is written mainly in TypeScript.

Yes. CoAI is open source, so the software itself is free to use.

Can I self-host CoAI?

Yes. CoAI can be self-hosted on your own server or infrastructure.

What are some alternatives to CoAI?

Similar open-source tools in the AI Infrastructure category include GPT-Load, LLM Gateway and Obot. SaaS products in the same category include AllToken, AnyRouter and Costbase.

Is CoAI actively maintained?

The most recent commit to CoAI was on 12 March 2026, and the latest release is v4.0.0, published on 23 October 2025. The project has 9.3k stars on GitHub.

CoAI: open-source AI infrastructure

About CoAI

CoAI, also written CoAI.Dev, is a self-hosted AI platform aimed at site owners who want to offer chat and model access to their own users. It bundles a chat web app with an API gateway in the style of One API, so a single deployment can serve people through a polished interface while routing requests to many model providers behind the scenes.

The gateway side supports more than 200 models from over 35 providers, including OpenAI, Anthropic and Gemini, with priority-based routing, load balancing, cost management and a built-in model cache. On the product side it adds an admin dashboard, credit and subscription billing, multi-device conversation sync with link sharing, a model market with presets, file parsing, web search, Markdown rendering with LaTeX and Mermaid, and text-to-image support.

CoAI is written with a Go backend and a React and TypeScript front end, and it is released under the Apache-2.0 licence. It is deployed on your own servers using Docker or other guides, with documentation and a Discord community. It suits teams and small businesses that want to resell or share access to language models with billing and administration included.

Key features

Unified gateway for many LLM providers
Priority-based routing and load balancing
Credit and subscription billing system
Admin dashboard with cost management
Cross-device conversation sync and sharing
File parsing and web search

Good fit for

→Running a shared AI chat service for a team or community
→Reselling model access with billing
→Consolidating several LLM API keys behind one gateway

Built with: TypeScript; Go; React
Tags: llm-gateway; ai-chat; multi-tenant; billing; openai; api-gateway; go; react; self-hosted

CoAI: questions and answers

What is CoAI used for?: CoAI is a multi-tenant AI platform that combines an LLM gateway for many providers with a chat interface, admin dashboard and credit or subscription billing. It is a good fit for running a shared AI chat service for a team or community, reselling model access with billing and consolidating several LLM API keys behind one gateway.
Is CoAI open source?: Yes. CoAI is open source under the Apache-2.0 licence. Its source code is on GitHub at coaidev/coai and is written mainly in TypeScript.
Is CoAI free?: Yes. CoAI is open source, so the software itself is free to use.
Can I self-host CoAI?: Yes. CoAI can be self-hosted on your own server or infrastructure.
What are some alternatives to CoAI?: Similar open-source tools in the AI Infrastructure category include GPT-Load, LLM Gateway and Obot. SaaS products in the same category include AllToken, AnyRouter and Costbase.
Is CoAI actively maintained?: The most recent commit to CoAI was on 12 March 2026, and the latest release is v4.0.0, published on 23 October 2025. The project has 9.3k stars on GitHub.