Llava-next

LLaVA-NeXT is an advanced, open-source multi-modal large language model architecture. It builds upon the original LLaVA framework, introducing enhanced visual reasoning, superior high-resolution image understanding, and stronger video processing capabilities.