| ▲ | jbs789 14 days ago | |
Source? | ||
| ▲ | dachworker 14 days ago | parent | next [-] | |
More informative than the article: https://xcancel.com/blelbach/status/1902113767066103949 cuTile seems to be the NVIDIA answer to OpenAI Triton. | ||
| ▲ | macksd 14 days ago | parent | prev | next [-] | |
It seems to pretty much be about these packages: https://nvidia.github.io/cuda-python/cuda-core/latest/ https://developer.nvidia.com/nvmath-python | ||
| ▲ | pjmlp 14 days ago | parent | prev [-] | |
The plethora of packages, including DSLs for compute and MLIR. https://developer.nvidia.com/how-to-cuda-python And "Zero to Hero: Programming Nvidia Hopper Tensor Core with MLIR's NVGPU Dialect" from 2024 EuroLLVM. | ||