apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-23 16:44:22 +08:00
Code Issues Actions 5 Packages Projects Releases Wiki Activity

[NewFeture]add ep rollout model init and update/clear ep buffer (#4039)

Browse Source 
* fix gid

* merge

* fix test

* fix bug

* fix

* fix ci
This commit is contained in:
gaoziyuan
2025-09-17 20:24:53 +08:00
committed by GitHub
parent 0d3a57a2c6
commit 896e3bb606
12 changed files with 348 additions and 293 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
custom_ops/gpu_ops/moe/ep_moe_expert_dispatch.cu
View File

@@ -33,6 +33,11 @@
__VA_ARGS__ \
break; \
} \
case 3: { \
constexpr size_t NUM_EXPERTS_PER_RANK = 3; \
__VA_ARGS__ \
break; \
} \
case 6: { \
constexpr size_t NUM_EXPERTS_PER_RANK = 6; \
__VA_ARGS__ \