CODA: GPU Kernels as GEMM-plus-Epilogue Programs

CODA is a GPU kernel abstraction that expresses Transformer operators as GEMM-plus-epilogue programs, fusing normalization, activations, residual updates, and reductions into the GEMM output tile before it is written to global memory, combining framework-level productivity with hardware-level efficiency. CODA is built on CUTLASS CuTeDSL and targets NVIDIA Hopper (H100) GPUs.

Updates

• July 19, 2026. Released v0.2.
• June 23, 2026. We are restructuring CODA. For legacy version, please check v0.1 tag.

Installation

pip install coda-kernels

Or from source:

git clone https://github.com/open-lm-engine/coda-kernels.git
cd coda-kernels
pip install -e .

Functional API

• linear_swiglu
• linear_cross_entropy
• linear_cross_entropy_forward
• linear_qknorm_rope