Translated data: A Zhejiang University Chu Kochen Honors College alumnus has collaborated with Microsoft Research to launch the multimodal model LLaVA, challenging GPT-4V. LLaVA has performed exceptionally well on 11 test datasets, garnering over 6,000 stars. The model's comprehensive capabilities exceed 85% of GPT-4V's level. The open-source code, model, and training data for LLaVA are now available for use.