apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-12-24 13:28:13 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[Feature] support eplb in api_server (#4782)

Browse Source 
* support eplb in api_server

* update code

* add eplb test case

* update eplb

* support tp+dp eplb

* update test cese

* update code

* update code

* fix bug

* update copilot review

* update test case name
This commit is contained in:
kevin
2025-11-24 20:22:29 +08:00
committed by GitHub
parent d5bd64336a
commit 8e4e3ff510
25 changed files with 2102 additions and 421 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
fastdeploy/model_executor/models/ernie4_5_moe.py
View File

@@ -368,7 +368,7 @@ class Ernie4_5_Model(nn.Layer):
fd_config.model_config.pretrained_config.prefix_name = "ernie"
self.fd_config = fd_config
self.redundant_table_manger = None
if fd_config.model_config.enable_redundant_experts is True:
if fd_config.eplb_config.enable_eplb is True:
self.redundant_table_manger = RedundantExpertManger(
n_routed_experts=fd_config.model_config.moe_num_experts,
num_hidden_layers=fd_config.model_config.num_hidden_layers,