3 people e1f3904799 [kernel] sm_100a CUDA backward for VSA block-sparse attention, 128-token blocks (#1866) ...

Co-authored-by: Pengcheng Li <pengchengl@ptyche0203.ptyche.clusters.nvidia.com>
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>

2026-09-29 16:57:22 -07:00

..

triton_kernels

[bugfix] Fix dead NVFP4 QAT plumbing: bench imports, int8 error string, smooth_q crash (#1723)

2026-09-24 17:09:33 -07:00

__init__.py

[misc] yapf pass under CI's interpreter (3.12) + pin hook language_version (#1702)

2026-08-08 22:24:17 -07:00

block_sparse_attn_256.py

[perf] Add FA4 CuTe backward support for VSA-256 (#1639)

2026-08-19 11:46:49 -07:00

block_sparse_attn_bwd_sm100a.py

[kernel] sm_100a CUDA backward for VSA block-sparse attention, 128-token blocks (#1866)

2026-09-29 16:57:22 -07:00

block_sparse_attn_cute_fwd.py

[perf] Add FA4 CuTe backward support for VSA-256 (#1639)

2026-08-19 11:46:49 -07:00

block_sparse_attn_sm100a.py

[kernel] sm_100a CUDA backward for VSA block-sparse attention, 128-token blocks (#1866)

2026-09-29 16:57:22 -07:00

block_sparse_attn_varlen.py

[misc] yapf pass under CI's interpreter (3.12) + pin hook language_version (#1702)

2026-08-08 22:24:17 -07:00

block_sparse_attn.py

[kernel] sm_100a CUDA backward for VSA block-sparse attention, 128-token blocks (#1866)

2026-09-29 16:57:22 -07:00

comm_ops.py

[perf] Add fused NVLink all-to-all for Ulysses (#1740)

2026-08-22 23:06:24 -07:00

ops.py

[perf] Add FA4 CuTe backward support for VSA-256 (#1639)

2026-08-19 11:46:49 -07:00

turbodiffusion_ops.py

[misc] yapf pass under CI's interpreter (3.12) + pin hook language_version (#1702)

2026-08-08 22:24:17 -07:00

version.py

[chore] release fastvideo-kernel 0.3.5 (#1775)

2026-08-27 23:18:59 -07:00

vmoba.py

[misc] yapf pass under CI's interpreter (3.12) + pin hook language_version (#1702)

2026-08-08 22:24:17 -07:00

vsa_utils.py

[feat] Add native SM103a VSA support (#1812)

2026-09-03 15:00:28 -07:00