Gregory Shtrasberg 56b325e977

[ROCm][AMD][Model]Adding alibi slopes support in ROCm triton flash attention and naive flash attention (#6043) ...

Co-authored-by: Hongxia Yang <62075498+hongxiayang@users.noreply.github.com>

2024-07-03 22:19:38 -07:00

..

backends

[ROCm][AMD][Model]Adding alibi slopes support in ROCm triton flash attention and naive flash attention (#6043)

2024-07-03 22:19:38 -07:00

ops

[hardware][misc] introduce platform abstraction (#6080)

2024-07-02 20:12:22 -07:00

__init__.py

[Core][2/N] Model runner refactoring part 2. Combine prepare prefill / decode to a single API (#4681)

2024-05-15 14:00:10 +09:00

layer.py

[Bugfix] Only add Attention.kv_scale if kv cache quantization is enabled (#5936)

2024-06-28 21:12:40 +00:00

selector.py

[Kernel] Flashinfer for prefill & decode, with Cudagraph support for decode (#4628)

2024-06-28 15:28:49 -07:00