Skip to content

fedorn/telebrain

 
 

Repository files navigation

Telebrain - Telegram Desktop with AI Copilot

Telebrain is a fork of Telegram Desktop with an integrated AI Copilot that brings powerful AI assistance directly into your Telegram experience. Chat with AI in the context of your conversation to get contextually-aware responses. The Completions API endpoint is fully configurable, allowing you to connect to most cloud LLM providers (OpenAI, Anthropic, DeepSeek) or run against a local model (via LM Studio or similar). This repository contains the complete source code and build instructions, based on the official Telegram messenger desktop client.

aznavour

The source code is published under GPLv3 with OpenSSL exception, the license is available here.

Support for custom Completions API endpoint, API key, and model

config

Tested endpoints (Base URL) and models

  • OpenAI - default: https://api.openai.com/v1, models gpt-5 (default), gpt-4.1, gpt-4o - requires API key
  • Anthropic: https://api.anthropic.com/v1, models claude-sonnet-4-20250514, claude-opus-4-1-20250805 - requires API key
  • DeepSeek: https://api.deepseek.com, models deepseek-chat, deepseek-reasoner - requires API key
  • LM Studio: http://localhost:1234/v1, model google/gemma-3-12b, openai/gpt-oss-20b (prints reasoning tokens without proper formatting) - doesn't require API key

Third-party

Build instructions

About

Telegram Desktop with AI Copilot

Resources

License

Contributing

Stars

Watchers

Forks

Packages

 
 
 

Languages

  • C++ 97.4%
  • Objective-C++ 0.9%
  • CMake 0.6%
  • Python 0.4%
  • CSS 0.2%
  • Shell 0.1%
  • Other 0.4%