中文版README

👋 Join our Wechat · 💡Try CogVLM2 Online 💡Try CogVLM2-Video Online

📍Experience the larger-scale CogVLM model on the ZhipuAI Open Platform.

• 🔥 News: 2024/8/30: The CogVLM2 paper has been published on arXiv.
• 🔥 News: 2024/7/12: We have released CogVLM2-Video online web demo, welcome to experience it.
• 🔥 News: 2024/7/8: We released the video understanding version of the CogVLM2 model, the CogVLM2-Video model. By extracting keyframes, it can interpret continuous images. The model can support videos of up to 1 minute. See more in our blog.
• 🔥 News: 2024/6/8:We release CogVLM2 TGI Weight, which is a model can be inferred in TGI. See Inference Code in here
• 🔥 News: 2024/6/5:We release GLM-4V-9B, which use the same data and training recipes as CogVLM2 but with GLM-9B as the language backbone. We removed visual experts to reduce the model size to 13B. More details at GLM-4 repo.
• 🔥 News: 2024/5/24: We have released the Int4 version model, which requires only 16GB of video memory for inference. You can also run on-the-fly int4 version by passing --quant 4.
• 🔥 News: 2024/5/20: We released the next generation model CogVLM2, which is based on llama3-8b and is equivalent (or better) to GPT-4V in most cases ! Welcome to download!

We launch a new generation of CogVLM2 series of models and open source two models based on Meta-Llama-3-8B-Instruct. Compared with the previous generation of CogVLM open source models, the CogVLM2 series of open source models have the following improvements:

1. Significant improvements in many benchmarks such as TextVQA, DocVQA.
2. Support 8K content length.
3. Support image resolution up to 1344 * 1344.
4. Provide an open source model version that supports both Chinese and English.

You can see the details of the CogVLM2 family of open source models in the table below:

Our open source models have achieved good results in many lists compared to the previous generation of CogVLM open source models. Its excellent performance can compete with some non-open source models, as shown in the table below:

All reviews were obtained without using any external OCR tools ("pixel only").

CogVLM2-Video achieves state-of-the-art performance on multiple video question answering tasks. The following diagram shows the performance of CogVLM2-Video on the MVBench, VideoChatGPT-Bench and Zero-shot VideoQA datasets (MSVD-QA, MSRVTT-QA, ActivityNet-QA). Where VCG-* refers to the VideoChatGPTBench, ZS-* refers to Zero-Shot VideoQA datasets and MV-* refers to main categories in the MVBench.

Performance on VideoChatGPT-Bench and Zero-shot VideoQA dataset:

Performance on MVBench dataset:

This open source repos will help developers to quickly get started with the basic calling methods of the CogVLM2 open source model, fine-tuning examples, OpenAI API format calling examples, etc. The specific project structure is as follows, you can click to enter the corresponding tutorial link:

• CLI demo, inference CogVLM2 model.
• CLI demo, inference CogVLM2 model using multiple GPUs.
• Web demo, provided by chainlit.
• API server, in OpenAI format.
• Int4 can be easily enabled with --quant 4, memory usage is 16GB.

• peft framework's efficient fine-tuning example.

• CLI demo, inference CogVLM2-Video model.
• Int4 can be easily enabled with --quant 4, with 16GB memory usage.
• Restful API server.
• Gradio demo.

In addition to the official inference code, you can also refer to the following community-provided inference solutions:

• xinference

This model is released under the CogVLM2 CogVLM2 LICENSE. For models built with Meta Llama 3, please also adhere to the LLAMA3_LICENSE.

If you find our work helpful, please consider citing the following papers

@article{hong2024cogvlm2,
  title={CogVLM2: Visual Language Models for Image and Video Understanding},
  author={Hong, Wenyi and Wang, Weihan and Ding, Ming and Yu, Wenmeng and Lv, Qingsong and Wang, Yan and Cheng, Yean and Huang, Shiyu and Ji, Junhui and Xue, Zhao and others},
  journal={arXiv preprint arXiv:2408.16500},
  year={2024}
}

@misc{wang2023cogvlm,
      title={CogVLM: Visual Expert for Pretrained Language Models}, 
      author={Weihan Wang and Qingsong Lv and Wenmeng Yu and Wenyi Hong and Ji Qi and Yan Wang and Junhui Ji and Zhuoyi Yang and Lei Zhao and Xixuan Song and Jiazheng Xu and Bin Xu and Juanzi Li and Yuxiao Dong and Ming Ding and Jie Tang},
      year={2023},
      eprint={2311.03079},
      archivePrefix={arXiv},
      primaryClass={cs.CV}
}