alexzms be548a78ea [feat] VSA-256 fastpath on Blackwell via FA4 CuTe block-sparse attention (#1354)

2026-05-26 12:58:29 -07:00

..

triton_kernels

[Perf] Skip bool-mask round-trip in block-sparse VSA attention (#1243)

2026-04-26 15:14:37 -07:00

__init__.py

[feat] VSA-256 fastpath on Blackwell via FA4 CuTe block-sparse attention (#1354)

2026-05-26 12:58:29 -07:00

block_sparse_attn_256.py

[feat] VSA-256 fastpath on Blackwell via FA4 CuTe block-sparse attention (#1354)

2026-05-26 12:58:29 -07:00

block_sparse_attn_cute_fwd.py

[feat] VSA-256 fastpath on Blackwell via FA4 CuTe block-sparse attention (#1354)

2026-05-26 12:58:29 -07:00

block_sparse_attn.py

[feat] VSA-256 fastpath on Blackwell via FA4 CuTe block-sparse attention (#1354)

2026-05-26 12:58:29 -07:00

ops.py

[feat] VSA-256 fastpath on Blackwell via FA4 CuTe block-sparse attention (#1354)

2026-05-26 12:58:29 -07:00

turbodiffusion_ops.py

[kernel] add turbodiffusion kernels (#972)

2025-12-30 04:23:10 -06:00

version.py

[bugfix] fastvideo-kernel: fix VSA Triton padding NaNs and support q/kv length mismatch (#1094)

2026-02-13 12:39:49 -08:00

vmoba.py

[misc]: standardize install instructions on uv pip install (#1279)

2026-05-02 12:45:50 -07:00