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

fix bug (#4679)

Browse Source
This commit is contained in:
kevin
2025-10-31 14:59:18 +08:00
committed by GitHub
parent 0f75b62de2
commit 096d87d335
1 changed files with 4 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
fastdeploy/config.py
View File

@@ -1556,6 +1556,10 @@ class FDConfig:
self.graph_opt_config.use_cudagraph = False
logger.info(f"CUDAGraph only support on GPU, current device type is {self.device_config.device_type}!")
if self.model_config.enable_mm and self.graph_opt_config.use_cudagraph:
self.cache_config.enable_prefix_caching = False
logger.info("Multi-modal models do not support prefix caching when using CUDAGraph!")
if self.scheduler_config.splitwise_role == "mixed":
self.model_config.moe_phase = MoEPhase(phase="prefill")
elif self.scheduler_config.splitwise_role == "prefill":