mirror of https://github.com/tile-ai/tilelang.git synced 2026-10-25 07:14:16 +08:00

Files

T

Lei Wang 32060ecda2 [Enhancement] Improve flashattn function in example_gqa_decode.py (#329 )

- Added a manual seed for reproducibility in PyTorch.
- Refactored local variable allocations for better memory management.
- Enhanced parallel processing in the flashattn function to improve performance.
- Updated layout annotations for clarity and efficiency.

These changes optimize the flash attention mechanism and ensure consistent behavior across runs.

2025-04-04 12:25:13 +08:00

example_gqa_decode.py

[Enhancement] Improve flashattn function in example_gqa_decode.py (#329 )

2025-04-04 12:25:13 +08:00

example_mha_inference.py

[Refactor] Deprecated T.Buffer as arguments and rename related calls into T.Tensor (#281 )

2025-03-26 13:42:52 +08:00

example_sparse_gqa_decode_varlen.py

[Dynamic Symbolic] Adaptively vectorize with different condition expressions (#326 )

2025-04-04 12:03:07 +08:00

heuristic.py

[Dynamic Symbolic] Adaptively vectorize with different condition expressions (#326 )

2025-04-04 12:03:07 +08:00

README.md

[Doc] Remove unnecessary layout annotation (#49 )

2025-01-25 17:05:05 +08:00

README.md

Flash Decoding