paddleocr-vl推理出现内存问题:malloc_consolidate(): invalid chunk size #16795
Answered
by
wcelll
jackliu1111
asked this question in
Q&A
Replies: 1 comment
-
meet the same issue,on nvidia T4
|
Beta Was this translation helpful? Give feedback.
0 replies
Answer selected by
jackliu1111
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment

Uh oh!
There was an error while loading. Please reload this page.
-
问题描述
我在使用paddleocr-vl官方推荐的环境下使用A100进行推理,会出现malloc_consolidate(): invalid chunk size或malloc_consolidate(): unaligned fastbin chunk detected等问题。
环境复现
宿主机环境是centos linux7,
paddle环境配置流程:
调用方式为(模型已通过vllm的docker镜像部署):
pipeline = PaddleOCRVL(vl_rec_backend="vllm-server", vl_rec_server_url="http://10.106.24.12:8118/v1")
output = pipeline.predict("xxx.pdf")
截图

不知道这个内存问题是怎么导致的,有概率出现
Beta Was this translation helpful? Give feedback.
All reactions