Open source compiler for your whole ML repo (support python, CUDA C/C++) into sass and cuda binary (which are closed sourced by Nvidia); 3-40% faster than Openai Triton for Neural Network (primarily LLM) based training/inferencing
python linux sass compiler gpu modal cuda nvidia triton gpu-acceleration compiler-optimization nvidia-cuda ptx compiler-backend cuda-c a100 llm cuda-cpp h100
-
Updated
Nov 14, 2026 - Python