Qwen2-VL-7B-Instruct Video inference
Closed this issue · 1 comments
Describe the bug
What the bug is, and how to reproduce, better with screenshots(描述bug以及复现过程，最好有截图)
When I use the reasoning provided by Qwen2-VL for video description, the results can be returned normally.
When I use the reasoning provided by Swift for video description, I will encounter the following error message
Your hardware and system info
Write your system info like CUDA version/system/GPU/torch version here(在这里给出硬件信息和系统信息，如CUDA版本，系统，GPU型号和torch版本等)
torch 2.2.1+cu121
torch-tb-profiler 0.4.3
torchaudio 2.2.1+cu121
torchvision 0.17.1+cu121
modelscope 1.17.1
modelscope_studio 0.4.0.9
ms-swift 2.4.0.post1
Additional context
Add any other context about the problem here(在这里补充其他信息)
If torch < 2.4, only local_path is supported, or you need to update to torch >= 2.4.