How to use tsinghua-ee/video-SALMONN-2_plus_3B with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("output/audio_align_3b_250910/checkpoint-15255") model = PeftModel.from_pretrained(base_model, "tsinghua-ee/video-SALMONN-2_plus_3B")
nielsr