Deploy a private AI assistant inside your Telegram account that you can chat with using OpenAI, Claude, or Gemini.
Switch between different AI models mid-conversation using Telegram inline buttons without any code changes.
Run multiple separate Telegram bots from one Cloudflare Workers deployment for different teams or projects.
Add text-to-image generation to a Telegram bot and let users create images with a simple chat command.
Requires a Telegram bot token and at least one AI provider API key, Cloudflare Workers free tier is sufficient.
This project lets you run your own AI chatbot inside Telegram by deploying a small script to Cloudflare Workers, which is a serverless hosting platform that runs code without you needing a traditional server or domain name. The setup is described as a copy-and-paste operation, and you do not need to install anything on your local computer to get it running. The bot connects to a wide range of AI providers, including OpenAI, Anthropic, Google Gemini, Groq, Mistral, and several others, so you can switch between different AI models directly inside Telegram using on-screen buttons. You can also define a custom starting personality for the bot, which it will remember across conversations. Beyond Cloudflare Workers, the same codebase can be deployed on Vercel, Docker, or other platforms if you prefer a different hosting setup. The project supports running multiple separate Telegram bots from one deployment and includes a plugin system for adding custom behavior. Other features include streaming responses so replies appear as they are being generated, text-to-image generation, and support for multiple languages. The README links out to separate documentation files for deployment steps, configuration options, and the plugin system. The project is released under the MIT license.
← tbxark on gitmyhub — every repo by this author, as a profile.
Verify against the repo before relying on details.