Build compute kernels and load them from the Hub.
Hugging Face Kernels is a system that lets Python libraries and applications load compute kernels directly from the Hugging Face Hub. It is used by Python and PyTorch developers who need portable kernels that can support multiple versions and build configurations. The catalogue skills help agents work with the kernels system and its kernel-building tools.
Latest release v0.18.1 · 2 Nov 2026
1 file for Codex and OpenCode: kernels AGENTS.md — 1,090 tokens loaded in every session.
AGENTS.md A 1,090 tok These files are huggingface/kernels's own configuration — they tell Codex and OpenCode how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.