How to use video reasoning for video frames？

Hello author! Thank you very much for your open source and I am looking forward to trying your model for inference. However, I am currently facing a problem where I have extracted my video as frames. If I only have frames, can I still use video for inference? Or can we only use multi graph reasoning?
I have tried to put the frames in a list and input them into the model, but still got the error Number of images does not match the number of image tokens in the text
Looking forward to your reply! Wishing you all the best

The issue names no file, test, or entry point. Start by checking the documented inference interface for frame-list inputs and reproduce the reported image-token mismatch. Done means establishing whether extracted frames are supported and recording the required input format or a clear limitation.

Written by the indexing model from the issue text.