Liyun Zhu

icon indicating a website link https://tom-roujiang.github.io/liyun_zhu/

icon location Canberra, Australia icon location ANU CS Student | Computer Vision | Video Understanding | VLM | Reasoning

Results 1 comments of Liyun Zhu

Qwen3-VL-30B-A3B-Thinking无法进行输入视频

I also have the same problem when inferencing Qwen 3 VL with vllm. openai.InternalServerError: Error code: 500 - {'error': {'message': 'Expected reading 914 frames, but only loaded 912 frames from...