[Feature] Flash Attention to Support Attention Mask (#760)

Signed-off-by: Didan Deng <33117903+wtomin@users.noreply.github.com>
This commit is contained in:
Didan Deng authored and GitHub committed 2026-01-16 11:01:11 +08:00
1 parent 35f994e255
commit 43ec7f701b
6 files changed
+566 -12

No files matched your search

@@ -0,0 +1,13 @@
# SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
"""
Utils for attention backends.
"""
from vllm_omni.diffusion.attention.backends.utils.fa import _pad_input, _unpad_input, _upad_input
__all__ = [
"_pad_input",
"_unpad_input",
"_upad_input",
]