Kernels are written in the CuTe-DSL.

pip install quack-kernels

• H100 or B200 GPU
• CUDA toolkit 12.9+
• Python 3.12

• 🦆 RMSNorm forward + backward
• 🦆 Softmax forward + backward
• 🦆 Cross entropy forward + backward
• 🦆 Layernorm forward

Upcoming:

• 🦆 Rotary forward + backward

from quack import rmsnorm, softmax, cross_entropy

[2025-07-10] We have a comprehensive blogpost on how to get memory-bound kernels to speed-of-light, right in the comfort of Python thanks to the CuTe-DSL.

See our blogpost for the details.

To set up the development environment:

pip install -e '.[dev]'
pre-commit install