OpenGVLabinAI AdvancesOpen-source vision-language model now comparable to GPT-4VInternVL 1.5 VS GPT4V. Here are some real cases.May 2May 2
OpenGVLabinAI AdvancesDoes Your Multi-model LLM Truly See The Diagrams In Visual Math Problems?If you want to use an AI model to solve math problems, you need to make sure which one understands diagrams.Apr 17Apr 17
OpenGVLabinAI AdvancesInternVid: Video-Text Dataset to Empowering Video Creation and UnderstandingA large-scale video-text dataset contains over 7 million videos.Apr 3Apr 3
OpenGVLabinAI AdvancesVideoMamba: State Space Model for Efficient Video UnderstandingBetter, faster, cheaper method for Video understanding with AIMar 27Mar 27
OpenGVLabinAI AdvancesThe All-Seeing Project: Towards Panoptic Visual Recognization and General Relation Comprehension…Mar 26Mar 26
OpenGVLabOmniQuant: Calibrated Quantization for LLMs, Has been Integrated with commercial APPAn open-source, efficient LLM Model quantization method.Mar 12Mar 12