lmms-lab

/

LLaVA-NeXT-Video-7B-Qwen2

like 2

Yes. In terms of video. It is a video specific model

Thanks! Is inference the same as llava-OneVision? I.e. all the same tokens, dimensions etc?