Setting up the MiroFlow environment requires three main steps:
- Cloning the repository and preparing the Python environment: using uv and Python ≥ version 3.12
- Configuring Environment Variables: 7 core API keys including HF_TOKEN (dataset), OPENROUTER_API_KEY (master model), ANTHROPIC_API_KEY (visualization tool), etc. are required.
- Optional E2B sandbox configuration: Building Reproducible Linux Environments with npm and Docker
Key Note: All API keys should be stored in the .env file and it is recommended to copy the changes from the template file. If you use other LLM as the master model, you also need to provide the corresponding API key.
This answer comes from the articleMiroFlow: a framework for building, managing and scaling AI intelligencesThe