Lightweight Python Telegram bot for LLM chatting with user separation and multiple session support bot.py requirements.txt .env.example What it supports: LLM chat over Telegram Separate session lists per Telegram user Independent message history per session Session switching Session creation Session renaming Current session lookup Commands: /start /help /new [name] /sessions /use <session_id> /rename <session_id> /current Implementation notes: Uses sqlite3 for lightweight persistence Stores: users sessions messages Uses an OpenAI-compatible backend via env vars: TELEGRAM_BOT_TOKEN OPENAI_API_KEY OPENAI_MODEL optional OPENAI_BASE_URL Run steps: Install deps: pip install -r requirements.txt Copy env template and fill values: cp .env.example .env Start bot: python3 bot.py Made by Codex, GPT-5.4