# Visual question answering enhancement

Spatial-LLaVA-7B is a multimodal model fine-tuned based on the LLaVA model, focusing on improving the ability of spatial relationship reasoning and suitable for multimodal research and chatbot development.