Warp on ROCm documentation#
2026-10-30
1 min read time
Warp is a Python framework for writing high-performance simulation and graphics code. Warp takes regular Python functions and JIT compiles them to efficient kernel code that can run on the CPU or GPU. It comes with a rich set of primitives that make it easy to write programs for physics simulation, perception, robotics, and geometry processing.
Accelerate physics simulation, robotics, and spatial computing workloads with Warp on ROCm for AMD Instinct GPUs, writing GPU kernels in regular Python that are JIT-compiled to efficient HIP code.
Warp on ROCm adds HIP support so that Warp kernels run on AMD Instinct MI325X, MI350X, and MI355X GPUs. Warp kernels are differentiable and can be used as part of machine-learning pipelines with frameworks such as PyTorch.
Warp is part of the AMD Simulation toolkit.
The Warp public repository is located at AMD-Ecosystem/warp.
To contribute to the documentation, refer to Contributing to Warp.
You can find licensing information on the Licensing page.