generated at
LLaVA
https://github.com/haotian-liu/LLaVAhaotian-liu/LLaVA
> マルチモーダル GPT-4 レベル機能を目指して構築された大規模な言語および視覚アシスタント
https://llava-vl.github.ioProject
https://llava.hliu.ccDemo
https://github.com/haotian-liu/LLaVA/blob/main/docs/MODEL_ZOO.mdModel Zoo
CLIP ViT/L-14とVicunaを接続する

Llama 2対応
https://huggingface.co/liuhaotian/llava-llama-2-13b-chat-lightning-preview

LLaVA-1.5
https://arxiv.org/abs/2310.03744Improved Baselines with Visual Instruction Tuning

LLaVA-Plus
https://arxiv.org/abs/2311.05437LLaVA-Plus: Learning to Use Tools for Creating Multimodal Agents