Runway introduces Praxis-1 world action model for robotics
Runway AI Inc. introduced Praxis-1, an open-weight robotics policy model built on its large-scale video pretraining. The model learns from third‑person video and can adapt to different robot embodiments with light fine‑tuning.
Key points
- Runway AI Inc. unveiled Praxis‑1, an open‑weight robotics policy model built on its large‑scale video pretraining.
- Early partners Noble Machines, Standard Bots and Ultra are testing Praxis‑1 on their own hardware.
- Runway claims its world‑model simulation predicts real‑world robot performance with 0.95 correlation.
The company says the model has been tested with early partners Noble Machines, Standard Bots, and Ultra, and that simulating policies inside its world model predicts real‑world results with a 0.95 correlation. Praxis‑1 covers tasks from pick‑and‑place to deformable‑object manipulation, and Runway plans to release it publicly in the coming months with open weights.
Runway positions the model as a way to leverage abundant video data for robot training, arguing that open American models are needed for U.S. leadership in physical AI. The company will continue testing and evaluating the model with partners before full launch.
Model page: Praxis-1 →
The story so far
2 episodes →- Runway introduces Praxis-1 world action model for roboticsthis story
Runway introduces Praxis-1 world action model for robotics
The Robot Report · 2 October 2026
Loading the full article…
This text was published by The Robot Report and written by Brianna Wessling. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Generative AI & Models
All →- PewDiePie releases Ajax, an uncensored Qwen3.5-9B model for home PCs · 1 src
- Anthropic commits $518 billion in AI infrastructure spending over next decade · 5 src
- Black Forest Labs releases Flux 3 Image with multi-step editing · 1 src
- Perplexity AI releases decision model pplx-decider-v1-27b fine‑tuned from Qwen3.8-27B · 1 src
- Author finds 'just right' prompt range speeds GPT-6 Luna High · 1 src
Comments
via GitHub Discussions