黄仁勋的物理AI ChatGPT时刻,正被这家中国公司的“流式多模态”接棒 - 智东西
Quick Answer
Om AI联汇 introduces a groundbreaking 'streaming multimodal' model architecture, enabling real-time perception and action in physical AI applications.
Quick Take
Om AI联汇 introduces a groundbreaking 'streaming multimodal' model architecture, enabling real-time perception and action in applications. This approach emphasizes continuous observation and decision-making, essential for robotics and autonomous systems, marking a significant shift from traditional processing methods.
Key Points
- Om AI联汇's VLX series achieves millisecond-level real-time perception.
- The model enables continuous perception, precise localization, and action decision-making.
- General Vision Intelligence is crucial for scalable physical AI applications.
- Physical AI requires a complete intelligent system, not just a single model.
- The approach shifts focus from offline processing to real-time action in physical environments.
Article Excerpt
From source RSS / original summary快讯 头条 人工智能 芯东西 AIoT 云与智慧城市 机器人 VR/AR 手机通信 活动. 黄仁勋的物理AI ChatGPT时刻,正被这家中国公司的“流式多模态”接棒. 人工智能 智东西 智东西头条2026/07/01. **智东西(公众号:zhidxcom)** **作者 | 王涵** **编辑 | 漠影**. 在他看来,AI的演进可以分为四个阶段:**Perception AI、Generative AI、Agentic AI、**。 当模型能够理解质量、摩擦、惯性、动量守恒,AI才真正走出屏幕。 他同时指出,要让机器人理解物理世界,不能仅靠单一模型,而是需要**建立一整套智能系统**。. AI真正走向物理世界,机器人、无人机、安防摄像头、可穿戴设备这些场景,需要的不是回答问题,而是**持续工作**。 物理AI最重要的,也就是**主动执行**的能力。. **Om AI联汇CEO兼首席科学家赵天成博士**表示:“之前整个业内对通用视觉智能的关注度偏低,大家可能更关注一些可以看秀的表演或操作场景。
但**通用视觉**这个点是未来物理AI真正规模化应用落地必不可少的,而且可能是更加现实、更加直接的核心技术,会更广泛地应用到所有物理AI场景。 ”. 通用视觉智能(General Vision Intelligence),即模型能像人一样持续观察环境、精准定位目标、自主驱动行动,且这一切必须在端侧完成。. 这是**业界首次**提出 “流式多模态” 这一全新模型架构。 区别于传统模型“采集-上传-离线处理”的路径,VLX系列面向物理世界中持续涌入的视频流,实现**毫秒级实时感知**,并**首次**在端侧打通**“持续感知→精准定位→行动决策”**的完整闭环。. ## 一、三个模型、三层能力、一条链路. Om AI联汇的定义是三项核心能力:**持续感知**(无需人工
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from WebSearch (Tavily)
See more →WSJ: OpenAI is considering deep price reductions as competition ...
OpenAI is contemplating significant price cuts in response to competitive pressure from Anthropic, particularly due to the success of Claude Code in developer and coding workflows. This shift could affect pricing strategies in the AI market as companies vie for dominance in coding solutions.