Commit 9cc373f

authored

[Kernel][Amd] Add fp8 kv cache support for rocm custom paged attention (#8577)

1 parent 76515f3 commit 9cc373fCopy full SHA for 9cc373f

6 files changed

+246

-283

lines changed

csrc/rocm
tests/kernels
- test_attention.py
vllm
- _custom_ops.py
- attention/backends
  - rocm_flash_attn.py

6 files changed

+246

-283

lines changed

Comments

(0)