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

[XPU] xpu support mm prefix cache (#5356)

Browse Source 
Co-authored-by: ddchenhao66 <dhaochen163.com>
This commit is contained in:
ddchenhao66
2025-12-03 19:07:34 +08:00
committed by GitHub
parent a4bb3e9960
commit 4e8096bd0d
3 changed files with 184 additions and 51 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
fastdeploy/engine/args_utils.py
View File

@@ -1233,10 +1233,6 @@ class EngineArgs:
all_dict = asdict(self)
model_cfg = ModelConfig(all_dict)
# XPU currently disable prefix cache for VL model
if current_platform.is_xpu() and (self.enable_mm or model_cfg.enable_mm):
self.enable_prefix_caching = False
if not model_cfg.is_unified_ckpt and hasattr(model_cfg, "tensor_parallel_size"):
self.tensor_parallel_size = model_cfg.tensor_parallel_size