Papers accepted at NeurIPS 2024: MMStar, Prism, ShareGPT4Video
Paper accepted at ECCV 2024: ShareGPT4V
Papers accepted at CVPR 2024: Rein, FreeDrag
Mar 2025: MMStar ranked in Top 10 Most Influential NeurIPS 2024 Papers
Sep 2024: ShareGPT4V ranked in Top 5 Most Influential ECCV 2024 Papers
Aug 2024: Ranked 3rd on HuggingFace’s July 2024 Most Impactful Users leaderboard
Jun 2024: Released ShareGPT4Video project featuring 40K GPT4V-generated captions, 4.8M high-quality captions, a general video captioner, and the ShareGPT4Video-8B multimodal model
May 2024: Open-sourced Open-LLaVA-Next, a full implementation of LLaVA-NeXT series with all training data and checkpoints publicly available
Apr 2024: Released MMStar, an elite vision-indispensable multimodal benchmark
Nov 2023: Launched ShareGPT4V project with 100K GPT4-Vision captions, 1.2M high-quality captions, a general image captioner, and the ShareGPT4V-7B model