apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-05 00:33:03 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

Merge vl execution path into normal execution path (#2829)

Browse Source 
* merge vl model into gpu_model runner

Change-Id: I9f4691a3d5f135e8d72b1d58abcd15ef3aa3f2a6

* fix chinese

Change-Id: Ic7405109b984c21e076fb3b01ff6feb571d0119a

* fix the parse parameter

Change-Id: I4cd62ee87c06220af580d91e347145d4394917fe

* fix the bug in online_inference

Change-Id: Idb111bb2114e83017c4050b2a68cf039c6d3c559

* polish code

Change-Id: I7d4194102c2f1b0743b74fbd5fc284eb8ef4d17c
This commit is contained in:
Zero Rains
2025-07-15 22:20:03 +08:00
committed by GitHub
parent 5fc659b900
commit e7bcbbab52
9 changed files with 441 additions and 1732 deletions
Download Patch File Download Diff File Expand all files Collapse all files

21
fastdeploy/worker/output.py
View File

@@ -201,6 +201,27 @@ class ModelOutputData:
"""
accept_num: paddle.Tensor
"""
vl model enable to think
"""
enable_thinking: paddle.Tensor = None
"""
vl model think end id
"""
think_end_id: int = -1
"""
vl model need to think
"""
need_think_end: paddle.Tensor = None
"""
vl model reasoning index
"""
reasoning_index: paddle.Tensor = None
@dataclass
class ModelRunnerOutput: