* This dataset supports online use.Click here to jump.

The AI-CUDA-Engineer-Archive dataset is a collection of CUDA kernels generated by AI, designed to facilitate subsequent training of open source models and the development of better CUDA function modules. The dataset contains more than 30,000 CUDA kernels, all of which are generated by AI-driven CUDA engineers, of which more than 17,000 kernels have been verified to be correct, and about 50% of kernel performance is better than the native runtime of PyTorch.

It includes the PyTorch reference implementation, PyTorch/NCU/Clang-tidy performance analysis data, multiple kernel versions for each task, error information, and performance comparison data with PyTorch native and compiled runtimes.