kernels

The Kernel Hub allows Python libraries and applications to load compute kernels directly from the Hub. To support this kind of dynamic loading, Hub kernels differ from traditional Python kernel packages in that they are made to be:

Portable: a kernel can be loaded from paths outside PYTHONPATH.
Unique: multiple versions of the same kernel can be loaded in the same Python process.
Compatible: kernels must support all recent versions of Python and the different PyTorch build configurations (various CUDA versions and C++ ABIs). Furthermore, older C library versions must be supported.

🚀 Quick Start

Install the kernels package with pip (requires torch>=2.5 and CUDA):

pip install kernels

Here is how you would use the activation kernels from the Hugging Face Hub:

import torch

from kernels import get_kernel

# Download optimized kernels from the Hugging Face hub
activation = get_kernel("kernels-community/activation")

# Random tensor
x = torch.randn((10, 10), dtype=torch.float16, device="cuda")

# Run the kernel
y = torch.empty_like(x)
activation.gelu_fast(y, x)

print(y)

You can search for kernels on the Hub.

📚 Documentation

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

kernels-0.3.2.tar.gz (12.9 kB view details)

Uploaded Mar 20, 2025 Source

File details

Details for the file kernels-0.3.2.tar.gz.

File metadata

Download URL: kernels-0.3.2.tar.gz
Upload date: Mar 20, 2025
Size: 12.9 kB
Tags: Source
Uploaded using Trusted Publishing? No
Uploaded via: twine/6.1.0 CPython/3.11.9

File hashes

Hashes for kernels-0.3.2.tar.gz
Algorithm	Hash digest
SHA256	`0d1a37f84269bd473c1d8098a306e72c102712f8097020c3624caa44c7f98b82`
MD5	`7f4d641c363291a65eaf4973512e08dc`
BLAKE2b-256	`0880076dc081be0a9c92db8a114cd1ab1a65cd16abf136f013416f71572f22a7`