- VisionLLM: Large Language Model as Open-Ended Decoder for Vision-Centric Tasks
- VisionLLM v2: A Generalist Multimodal Large Language Model for Hundeds of Vision-Language Tasks
2024/06: We release VisionLLM v2, which is a generalist multimodal large language model to support hundres of vision-language tasks, covering visual understanding, perception and generation.