apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-09 10:30:37 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[XPU] support XPU VL model inference (#4030)

Browse Source 
* [XPU] support XPU VL model inference

* fix image op import and device check

* rebase develop

* fix perf
This commit is contained in:
Lucas
2025-09-25 14:34:15 +08:00
committed by GitHub
parent e36eccfdad
commit 87179cb744
18 changed files with 1300 additions and 146 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
fastdeploy/model_executor/forward_meta.py
View File

@@ -232,6 +232,8 @@ class XPUForwardMeta(ForwardMeta):
dec_batch: Optional[paddle.Tensor] = None
#
total_enc_len: Optional[paddle.Tensor] = None
# position embedding type in rope, supports 'NORMAL' or 'HALF_HEAD_DIM'
pos_emb_type: Optional[str] = "NORMAL"
@dataclass