meta-llama

/

Llama-3.2-11B-Vision-Instruct

like 331

Does this model support Multi Images？ if True，like this？

image1 = Image.open(url1)
image2 = Image.open(url2)

messages = [
{"role": "user", "content": [
{"type": "image"},
{"type": "image"},
{"type": "text", "text": "please describe these two images"}
]}
]
input_text = processor.apply_chat_template(messages, add_generation_prompt=True)
inputs = processor([image1, image2], input_text, return_tensors="pt").to(model.device)