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

[FDConfig] Turn on the CUDAGraph + MultiModel switch (#4512)

Browse Source
This commit is contained in:
RAM
2025-10-21 21:21:26 +08:00
committed by GitHub
parent 809c1ac7ec
commit d70aacfbdc
1 changed files with 3 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files

13
fastdeploy/config.py
View File

@@ -1510,14 +1510,12 @@ class FDConfig:
self.structured_outputs_config.guided_decoding_backend = "xgrammar"
# Adjustment GraphOptConfig
if (
(self.scheduler_config.splitwise_role != "mixed")
or (self.model_config is not None and self.model_config.enable_mm is True)
or (self.load_config is not None and self.load_config.dynamic_load_weight is True)
if (self.scheduler_config.splitwise_role != "mixed") or (
self.load_config is not None and self.load_config.dynamic_load_weight is True
):
self.graph_opt_config.use_cudagraph = False
logger.info(
"CUDAGraph does not support to be started together with MultiModel temporarily, but has been automatically closed!"
"CUDAGraph does not support to be started together with PD Disaggregation temporarily, but has been automatically closed!"
)
if self.load_config is not None and self.load_config.dynamic_load_weight is True:
self.graph_opt_config.graph_opt_level = 0
@@ -1632,11 +1630,6 @@ class FDConfig:
self.scheduler_config.check()
# Check graph optimization config
if self.graph_opt_config.use_cudagraph:
if self.model_config is not None:
assert (
self.model_config.enable_mm is not True
), "CUDAGraph cannot be applied to multimodal model temporarily"
if self.graph_opt_config.graph_opt_level > 0 or self.graph_opt_config.use_cudagraph:
if self.load_config is not None:
assert (