apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-05 16:48:03 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

native top_p_sampling (#2901)

Browse Source
This commit is contained in:
lifulll
2025-07-22 14:09:59 +08:00
committed by GitHub
parent 0eedbdaee0
commit 2c6a9e887e
14 changed files with 93 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
fastdeploy/model_executor/layers/attention/mla_attention_backend.py
View File

@@ -32,7 +32,7 @@ from fastdeploy.model_executor.layers.attention.ops import (
)
from fastdeploy.platforms import current_platform
if current_platform.is_cuda() and not current_platform.is_dcu():
if current_platform.is_cuda():
from fastdeploy.model_executor.ops.gpu import (
decode_mla_write_cache,
multi_head_latent_attention,