MIGraphX documentation

2025-11-20

1 min read time

Applies to Linux

MIGraphX is a graph compiler and inference engine for high performance machine learning model inference. It compiles trained models from end-to-end to optimize for inference performance on AMD hardware.

The MIGraphX public repository is located at ROCm/AMDMIGraphX.

You can integrate MIGraphX with PyTorch worflows by using the Torch-MIGraphX library. The public repository is located at ROCm/torch_migraphx.

Install

• MIGraphX on ROCm installation

• Torch-MIGraphX installation

Conceptual

• Deep learning compilation with MIGraphX

Reference

• MIGraphX user reference

  • MIGraphX C++ API reference

  • MIGraphX Python API reference

  • Supported ONNX Operators

• MIGraphX contributor reference

  • Environment variables

  • Develop for the MIGraphX code base

  • MIGraphX driver

Examples

• MIGraphX examples

To contribute to the documentation refer to Contributing to ROCm.

Licensing information can be found on the Licensing page.