Infer InternVL 1.5 without Image not work.

[Liac-li]

我用了 Hugging Face 中提供的代码实例

# single-round single-image conversation
question = "请详细描述图片" # Please describe the picture in detail
response = model.chat(tokenizer, pixel_values, question, generation_config)
print(question, response)

并且尝试把其中的 `pixel_values 赋值为 None。类似 llava 一样进行无图推理，但是不能运行。请问有别的方法让 InternVL1.5 进行无图推理吗？

---

I try to use InternVL1.5 to do inference without Image, like llava model. I copy the demo code from the HuggingFace Repository:

# single-round single-image conversation
question = "请详细描述图片" # Please describe the picture in detail
response = model.chat(tokenizer, pixel_values, question, generation_config)
print(question, response)`
并且尝试把其中的 `pixel_values

And try to give None to pixel_values, but the model is not support it. Is there any other method to make InternVL1.5 do inference without Image?