How to use fun-research/Video-LLaVA-Seg with Transformers:
# Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("fun-research/Video-LLaVA-Seg", dtype="auto")
