Yi-34B-VL 这张图会胡言乱语，为何

[lucasjinreal]

Reminder

• [X] I have searched the Github Discussion and issues and have not found anything similar to this.

Environment

OS: Ubuntu 20.04
Python: 3.10
PyTorch: 2.0.1+cu118
CUDA: 11.8

Current Behavior

this image caused model output wrong short unable to understand result.

Expected Behavior

shouldn't outputs meanless output

Steps to Reproduce

gdf

Anything Else?

[lucasjinreal]

[a2382625920]

Hi, is the problem solved, I had some similar problem today on Yi-VL-6B, the model can't output Chinese and generates it haphazardly. But the day before the problem, the model was still answering excellent!

[lucasjinreal]

No

[a2382625920]

不

OK, maybe there's something wrong with the model itself, or maybe there's something in the reasoning that doesn't fit the conditions

[Yimi81]

The open-source version of the VL model does not have good OCR capabilities, you can try our API