做视频生成的 Runway 出了个机器人模型,权重和代码全开源,Apache 2.0 随便用,还在 Nvidia 的 RoboLab-120 上拿了 World Action Model 类别第一。机器人方向的研究者和创业公司可以直接下载来微调试试。
#Praxis-1
Joint video-and-action prediction suggests Runway is steering its research toward world models that could serve robotics and simulation, not just media generation. Developers and researchers working on embodied AI or interactive video should read the technical report for architectural details unavailable in most commercial releases. It offers a direct comparison point against open world-model efforts that typically disclose far less about training design.
Running on one RTX 5090 means hobbyists and university labs can now deploy a foundation robotics model without data-centre budgets, a sharp contrast to closed offerings from physical-AI startups. Teams using Hugging Face's LeRobot stack can clone the checkpoint and follow the published setup instructions to get a SO-101 arm running the same day. Watch this space as community-dataset-trained models challenge proprietary robotics stacks.
An open-weight, Apache 2.0 robot policy model that runs on a single RTX 5090 lowers the entry bar for robotics labs and hardware startups that cannot afford proprietary stacks. Teams building manipulation systems can fine-tune it locally instead of licensing closed models like GPT-6 Astra, which Runway claims is 51 times costlier per task. Builders tracking video-to-action transfer, an emerging alternative to pure imitation learning, should benchmark it on their own hardware.
Apache 2.0 weights plus single-H100 qualification make Praxis-1 one of the more accessible open robot policies to self-host, with no cloud dependency for inference. Teams building robot stacks in ASEAN research labs or startups can prototype without multi-GPU budgets, and the OpenPI/Positronic serving paths mean existing manipulation pipelines can swap in the model with modest glue code. Anyone comparing open robot foundation models should weigh it against restricted-license alternatives.
An open-weights, Apache 2.0 robot control model removes the licensing barriers that usually come with robotics foundation models, letting teams self-host and modify freely. Robotics engineers and embodied-AI researchers can fine-tune Praxis-1 on their own hardware instead of licensing closed systems. Compared with closed offerings from rivals such as Physical Intelligence, this gives startups a low-cost starting point for building real-world robot behaviour.
Runway 把自家机器人模型 Praxis-1 和 GPT-6 Astra 放在同一个工作站上比,结果快了 15.6 倍,每个任务还便宜 51 倍。做机器人或自动化的开发者可以关注这种专用模型和通用 LLM 的选型对比。