ColabPilot

Bridge LLM chat UIs with Google Colab GPU execution. Run code blocks from ChatGPT, Claude, and Gemini directly in Colab.

Bridge your favorite AI chat with Google Colab's free GPU. ColabPilot detects code blocks in ChatGPT, Claude, and Gemini, runs them directly on your Colab notebook, and sends the output back to the chat all in one click. Supported Platforms - ChatGPT (chatgpt.com) - Claude (claude.ai) - Gemini (gemini.google.com) How It Works 1. Install the extension and open Google Colab 2. Run !pip install colabpilot then import colabpilot; colabpilot.connect() 3. Go to any supported AI chat and ask it to write code 4. Click "Run in Colab" on any code block — output is captured and sent back Features - One-click execution of Python and Bash code on Colab GPU - Auto-mode: automatically runs code and feeds results back for workflows - Persistent namespace: variables carry over between executions - Side panel with execution logs, queue status, and settings - No API keys needed: uses Colab's built-in RPC system Open Source ColabPilot is fully open source under the MIT License. GitHub: https://github.com/navaneethks/colabpilot Privacy No data is collected or sent to any server. All communication happens locally between your browser tabs via Chrome's messaging API.