首页开源帮助登录注册

trending/tilelang

mirror of https://github.com/tile-ai/tilelang.git synced 2026-11-09 09:46:50 +08:00

Files

T

History

Tong WU 0c7e74196e [Cleanup] Remove tilelang.disable_cache() calls from examples and tests (#1088 )

* [Cleanup] Remove `tilelang.disable_cache()` calls from example scripts

* lint

* lint

2025-10-21 16:34:30 +08:00

..

block_sparse_attn_triton.py

[Enhancement] Introduce flag to visualize shared memory merge plan (#496 )

2025-05-16 21:00:18 +08:00

example_tilelang_block_sparse_attn.py

[Refactor] Turn off ENABLE_FAST_MATH by default (#846 )

2025-09-18 16:07:13 +08:00

example_tilelang_sparse_gqa_decode_paged.py

[Cleanup] Remove tilelang.disable_cache() calls from examples and tests (#1088 )

2025-10-21 16:34:30 +08:00

example_tilelang_sparse_gqa_decode_varlen_indice.py

[Language] Recommend using T.dynamic instead of T.symbolic (#1076 )

2025-10-20 19:48:31 +08:00

example_tilelang_sparse_gqa_decode_varlen_mask.py

[Language] Recommend using T.dynamic instead of T.symbolic (#1076 )

2025-10-20 19:48:31 +08:00

example_triton_sparse_gqa_decode_varlen_indice.py

[Layout] Introduce a new layout inference mechanism (#699 )

2025-08-08 22:09:06 +08:00

example_triton_sparse_gqa_decode_varlen_mask.py

[Layout] Introduce a new layout inference mechanism (#699 )

2025-08-08 22:09:06 +08:00

heuristic.py

[Example] Add sparse gqa decode example (#332 )

2025-04-05 12:57:18 +08:00

README.md

[Example] Add paged block-sparse flash-decoding kernel (#638 )

2025-07-17 01:50:02 +08:00

test_example_blocksparse_attention.py

[CI]:Reduce test shapes to avoid OOM errors during CI. (#1060 )

2025-10-19 02:43:00 +08:00

README.md

Block-Sparse Flash-Attention

Tilelang implementation of block-sparse flash-attention kernels.

The kernels have been used in Rectified Sparse Attention and SeerAttention-R.