According to JD’s Blackboard Report dated June 22, JD has recently made its real-time video visual language interaction model, JoyAI-VL-Interaction, available as open-source software. This model represents the world’s inaugural full-stack open-source interaction framework and system, boasting native support from vLLM-Omni right from day zero. It propels large-scale AI models beyond mere 'question-and-answer' capabilities, ushering in an era of 'seeing-and-talking'. Leveraging this framework, developers can swiftly construct practical AI assistants capable of continuous observation, autonomous decision-making, and instantaneous responses.
