{"library":"kernels","type":"library","category":null,"description":"The `kernels` Python library, currently at version 0.13.0, enables other Python libraries and applications to dynamically load optimized compute kernels directly from the Hugging Face Kernel Hub. These kernels are designed to be portable, unique (multiple versions can coexist), and compatible across various Python and PyTorch configurations. The library supports a rapid release cycle, with frequent updates to add new features and improve performance.","language":"python","status":"active","version":"0.13.0","tags":["huggingface","machine-learning","compute-kernels","pytorch","cuda","gpu-acceleration"],"last_verified":"Sat May 23","install":[{"cmd":"pip install kernels","imports":["from kernels import get_kernel"]},{"cmd":"pip install kernels torch>=2.5","imports":[]}],"homepage":null,"github":null,"docs":null,"changelog":null,"pypi":"https://pypi.org/project/kernels/","npm":null,"openapi_spec":null,"status_page":null,"smithery":null,"compatibility":{"summary":{"python_range":"3.10–3.9","success_rate":75,"avg_install_s":41.1,"avg_import_s":1.43,"wheel_type":"wheel"},"url":"https://checklist.day/v1/registry/kernels/compatibility"}}