apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2026-04-23 00:17:25 +08:00
Code Issues Actions 19 Packages Projects Releases Wiki Activity

[XPU] refine fused moe (#4219)

Browse Source
This commit is contained in:
zhupengyang
2025-10-16 19:04:07 +08:00
committed by GitHub
parent 3bbe99eae7
commit 26ff2f8683
7 changed files with 354 additions and 585 deletions
Download Patch File Download Diff File Expand all files Collapse all files
fastdeploy/worker/worker_process.py
+3 -3
View File
@@ -404,9 +404,9 @@ class PaddleDisWorkerProc:
if num_blocks_local <= 0:
raise ValueError(
"The total number of blocks cannot be less than zero."
"Please increase gpu_memory_utilization"
"Or decrease max_num_batched_tokens(max model length) "
"The total number of blocks cannot be less than zero. "
"Please increase gpu_memory_utilization "
"Or decrease max_num_batched_tokens(max model length)."
)
if self.ranks > 1: