This project provides a versatile Telegram bot that integrates with multiple large language models (LLMs) and AI services, enabling users to interact with advanced AI capabilities directly within Telegram. It targets users seeking a unified interface for various AI models, including text generation, image creation, and web search, offering a feature-rich and customizable experience.

How It Works

The bot leverages a Python backend to asynchronously process messages and manage conversations across different LLMs. It supports multimodal inputs (voice, image, documents) and features a model grouping system for intuitive selection. A key design choice is its self-developed API request SDK, which avoids direct reliance on official SDKs, potentially offering greater flexibility and control over API interactions.

Quick Start & Requirements

• Install/Run: Deployment options include Docker, Zeabur, Replit, and fly.io. For local execution from source: git clone --recurse-submodules ..., pip install -r requirements.txt, python bot.py.
• Prerequisites: Python >= 3.10. API keys for desired LLM services (OpenAI, Groq, Gemini, Claude, etc.) are required. Google API keys are needed for Google Search.
• Links: Project Repository, Official Documentation

Highlighted Details

• Supports a wide array of LLMs: GPT-3.5/4/4o/o1, DALL·E 3, Claude 2.1/3/3.5, Gemini 1.5 Pro/Flash, Vertex AI, Groq, and more.
• Features multimodal capabilities for voice, image, and document (PDF, TXT, MD, Python) processing.
• Includes a plugin system for web search (DuckDuckGo, Google), URL summarization, ArXiv paper summarization, and code interpretation.
• Offers group chat topic modes, dialogue isolation, and customizable user preferences.

Maintenance & Community

The project has received sponsorship from several individuals and entities, indicating community interest and support. Development activity and updates are ongoing.

Licensing & Compatibility

• License: GPL-3.0 License.
• Compatibility: GPLv3 is a strong copyleft license. Derivative works must also be licensed under GPLv3. Commercial use or linking with closed-source software may require careful consideration of license obligations.

Limitations & Caveats

The project's reliance on external API keys means costs are associated with usage. Some features, like document Q&A, require enabling conversation history (PASS_HISTORY), which incurs additional costs and is not enabled by default. The README notes that some free deployment platforms like Koyeb can be unstable.