Repository navigation
AttributeError: module 'torch.cuda' has no attribute 'OutOfMemoryError' #1627
Description
When using the langchain-chatGLM project to run, the model does not report an error and can be used normally, but when it is changed to FastChat, the following error appears. What is the reason for this?
Environment
python=3.8.16,torch=1.12.1,
T4
model
chatGLM-6B-int4
Error
AttributeError: module 'torch.cuda' has no attribute 'OutOfMemoryError'
Activity
moonlightnoodles commentedon Jun 7, 2024
Has your problem been solved? I'm having the same issue
chessing234 commentedon Jul 16, 2026
PR up: detect CUDA OOM without referencing torch.cuda.OutOfMemoryError.
fix: handle CUDA OOM without torch.cuda.OutOfMemoryError.