Model description

InternVL3_5_Flash employ an additional router to use dynamic resolution of input images and this is not compatible with previous class "InternVLForConditionalGeneration". As a result, I would like to inquire whether it is possible to support the Flash version of InternVL3-5.

Open source status

• The model implementation is available
• The model weights are available

Provide useful links for the implementation

https://huggingface.co/OpenGVLab/InternVL3_5-8B-Flash/blob/main/modeling_internvl_chat.py