WarpGBM MCP Service

fastapi>=0.109.0 uvicorn[standard]>=0.27.0 httpx>=0.27.0 pydantic>=2.0.0 numpy>=1.24.0 joblib>=1.3.0 PyJWT>=2.8.0 python-multipart>=0.0.6 onnx>=1.15.0 onnxruntime>=1.16.0 skl2onnx>=1.16.0 slowapi>=0.1.9 # Model backends lightgbm>=4.0.0 scikit-learn>=1.3.0 # Fallback for testing warpgbm # GPU-accelerated gradient boosting (requires CUDA)