ZwGao

Please make sure these folder model_zoo/llama, and model_zoo/llama_7B_hf contain the correct checkpoint. You can try removing the folders and running the code again. The app will automatically download the checkpoint...

Please ensure that the file stylegan2-ffhq-config-f.pt is fully downloaded and keep the access port consistent with the port mapped in InternGPT/docker/InternGPT_CN/docker-compose.yml.

We will resume the service as soon as possible. Thank you for your attention.

> 其实我想实现的是，一个promt中包含5张图片，问题中包含一张图片，模型从ABCD四张图片中选出一个作为答案返回 您好，目前的internvl 2.0 已经支持多图,欢迎体验。[LINK](https://huggingface.co/OpenGVLab/InternVL2-8B#quick-start)

Hi, It is still necessary to pretrain and align the MLP first. A more efficient training solution will leave for future work

Thanks for your attention, we have released our fine-tuning code [LINK](https://github.com/OpenGVLab/InternVL/blob/v2.0.0/document/How_to_finetune_mini_internvl_chat_v1_5_on_a_custom_dataset.md)

请问现在这个问题解决了吗

We closed this issue because there was no response for a long time.

你好，请问现在问题修复了吗？internvl需要从源代码安装，感谢关注，另外AutoGPTQ backend没有支持internvl 模型，建议您使用LMDeploy。See [LINK](https://github.com/InternLM/lmdeploy/blob/main/docs/en/inference/vl_pipeline.md)

感谢关注，具体请参考swift实现。