# ConvNeXt Model Card

## Model details

**Model type:** ConvNeXt is an open-source visual encoder trained by fine-tuning LLM on multimodal caption and instruction-following data. The base model is: laion/CLIP-convnext_large_d_320.laion2B-s29B-b131K-ft-soup.

**Model date:** ConvLLaVA-ConvNeXt-1024 was trained in March 2024.

Paper or resources for more information: https://github.com/alibaba/conv-llava/

Where to send questions or comments about the model: https://github.com/alibaba/conv-llava/issues

## Intended use
**Primary intended uses:** The primary use of ConvLLaVA-ConvNeXt is research on large multimodal models and chatbots.

## Paper
arxiv.org/abs/2405.15738