| library_name: kernels | |
| license: other | |
| This is the repository card of MiniMaxAI/msa that has been pushed on the Hub. It was built to be used with the [`kernels` library](https://github.com/huggingface/kernels). This card was automatically generated. | |
| ## How to use | |
| ```python | |
| # make sure `kernels` is installed: `pip install -U kernels` | |
| from kernels import get_kernel | |
| kernel_module = get_kernel("MiniMaxAI/msa", version=0) | |
| sparse_atten_func = kernel_module.sparse_atten_func | |
| sparse_atten_func(...) | |
| ``` | |
| ## Available functions | |
| - `sparse_atten_func` | |
| - `sparse_atten_nvfp4_kv_func` | |
| - `sparse_decode_atten_func` | |
| - `SparseDecodePagedAttentionWrapper` | |
| - `fp4_indexer_block_scores` | |
| - `build_k2q_csr` | |
| - `SparseK2qCsrBuilderSm100` | |
| - `Nvfp4QuantizedTensor` | |
| - `quantize_bf16_to_nvfp4_128x4` | |
| - `quantize_kv_bf16_to_nvfp4_128x4` | |
| - `dequantize_nvfp4_128x4_to_bf16` | |
| - `swizzle_nvfp4_scale_to_128x4` | |
| - `nvfp4_global_scale_from_amax` | |
| ## Benchmarks | |
| No benchmark available yet. | |
| ## Source code | |
| Source code of this kernel originally comes from https://github.com/MiniMax-AI/MSA and it was repurposed for compatibility with `kernels`. | |