快讯
Liquid AI 发布 Open d1 系列开源多模态决策模型，覆盖 d1-3B 与 d1-omni-600M
Liquid AI 宣布发布 Open d1：d1 决策模型系列中的两个开源权重多模态模型。d1-3B 支持文本+视觉，d1-omni-600M 支持文本+图像或文本+音频。团队表示这些模型可在从 @nvidia DGX 数据中心到 RTX 工作站再到 Jetson 边缘设备上实现实时决策。转发者 Derya Unutmaz 称此类单次前向完成决策的模型对机器人、自主实验室、医疗设备和物理 AI 具有重大意义。
事件来源
查看原文
RT Derya Unutmaz, MD Super cool new decision models from @LiquidAI! These are like Jev, which is an emerging new class of AI models that make decisions in a single pass rather than generating tokens one by one. But LiquidAI Open d1 are both open-weight and multimodal decision models running locally, making decisions in milliseconds! The implications of decision models for robotics, autonomous laboratories, medical devices, and physical AI are staggering and very exciting! In particular, I very much like that they work with @nvidia Jetson at the edge, very relevant for robotics. This will be the bridge of AI to physical world, where every task series are decision making for the next action. Liquid AI: Today we release Open d1: two open-weight multimodal models in our d1 decision model family. > d1-3B: text + vision > d1-omni-600M: text + image or text + audio > Real-time decision making anywhere, from data centers such as @nvidia DGX to RTX workstations to Jetson at the edge.