jvlunteren 01a583fea4

[Kernel] Decouple Tile Size from Block Size in Triton Unified Attention Kernel (#21197) ...

Signed-off-by: Jan van Lunteren <jvl@zurich.ibm.com>

2025-09-18 14:27:01 +00:00

..

backends

[Bug] Fix is_flashmla_supported Check Error (#24774)

2025-09-15 20:10:55 -06:00

layers

Directly get max encoder len from VLLM config in V1 (#24866)

2025-09-16 17:52:31 +00:00

ops

[Kernel] Decouple Tile Size from Block Size in Triton Unified Attention Kernel (#21197)

2025-09-18 14:27:01 +00:00

utils

[Attention] FlashAttn MLA (#14258)

2025-09-04 02:47:59 -07:00

__init__.py

Remove duplicate entry in vllm.attention.__all__ (#23296)

2025-08-20 17:14:59 -07:00

layer.py

[XPU] Whisper model support on XPU Platform (#25123)

2025-09-18 04:30:10 +00:00

selector.py

[gpt-oss] Enable gpt-oss on ampere (#22714)

2025-08-12 03:21:44 -07:00