| Name | Last modified | Size | Description | |
|---|---|---|---|---|
| Parent Directory | - | |||
| torch_utils.h | 2026-06-05 13:22 | 3.2K | ||
| torch_bindings.cpp | 2026-06-05 13:22 | 35K | ||
| topk.cu | 2026-06-05 13:22 | 12K | ||
| sampler.cu | 2026-05-30 04:53 | 28K | ||
| quantization/ | 2026-05-30 04:53 | - | ||
| pos_encoding_kernels.cu | 2026-05-30 04:53 | 8.8K | ||
| persistent_topk.cuh | 2026-06-05 13:22 | 45K | ||
| permute_cols.cu | 2026-05-13 15:24 | 3.2K | ||
| ops.h | 2026-06-05 13:22 | 28K | ||
| nvfp4_kv_cache_kernels.cu | 2026-05-30 04:53 | 12K | ||
| moe/ | 2026-06-05 13:22 | - | ||
| minimax_reduce_rms_kernel.cu | 2026-06-05 13:22 | 32K | ||
| mamba/ | 2026-05-30 04:53 | - | ||
| layernorm_quant_kernels.cu | 2026-06-05 13:22 | 12K | ||
| layernorm_kernels.cu | 2026-06-05 13:22 | 12K | ||
| launch_bounds_utils.h | 2026-06-05 13:22 | 2.2K | ||
| fused_qknorm_rope_kernel.cu | 2026-06-05 13:22 | 34K | ||
| fused_deepseek_v4_qnorm_rope_kv_insert_kernel.cu | 2026-06-05 13:22 | 56K | ||
| fp32_router_gemm_entry.cu | 2026-05-30 04:53 | 5.0K | ||
| fp32_router_gemm.cu | 2026-05-30 04:53 | 7.2K | ||
| dsv3_fused_a_gemm.cu | 2026-05-13 15:24 | 27K | ||
| dispatch_utils.h | 2026-05-30 04:53 | 7.7K | ||
| cutlass_extensions/ | 2026-06-05 13:22 | - | ||
| custom_all_reduce.cu | 2026-06-05 13:22 | 7.6K | ||
| cuda_view.cu | 2026-06-05 13:22 | 3.0K | ||
| cuda_vec_utils.cuh | 2026-05-30 04:53 | 11K | ||
| cuda_utils_kernels.cu | 2026-06-05 13:22 | 1.0K | ||
| concat_mla_q.cuh | 2026-05-30 04:53 | 2.2K | ||
| cache_kernels_fused.cu | 2026-05-30 04:53 | 13K | ||
| cache_kernels.cu | 2026-05-30 04:53 | 70K | ||
| attention/ | 2026-05-30 04:53 | - | ||
| async_util.cuh | 2026-06-05 13:22 | 3.3K | ||
| activation_kernels.cu | 2026-05-30 04:53 | 30K | ||