Praxis-1 at a glance
Runway made Praxis-1, its 27B-parameter world action model for robots, downloadable on November 11, 2026, releasing the weights, inference and fine-tuning code under Apache 2.0 for commercial use, with checkpoints for single-arm and bimanual robots on Hugging Face and GitHub.
Ask your AI about this story
Use this story in your business: Runway releases Praxis-1 robot model as open weights under Apache 2.0 (Runway). Ask your AI to explain what happened, how it could affect your business and how to put it to work. It also weighs price, speed, intelligence and the lab behind it. The link to the story is included.
What changed with Praxis-1
Runway released Praxis-1 to everyone on November 11, 2026, publishing the weights and code of its world action model under the Apache 2.0 license, including for commercial use. Praxis-1 is a 27B-parameter mixture-of-experts diffusion transformer that predicts video and robot actions together, built on the large-scale video pretraining behind Runway's general world models and first introduced in September 2026. Three checkpoints are published (base, a DROID fine-tune for single-arm robots and a YAM fine-tune for bimanual robots), with inference and full and LoRA fine-tuning code on GitHub and Hugging Face with native LeRobot support, serving in vLLM-Omni and hosted inference on Reactor.
In Runway's November 11, 2026 announcement, the DROID fine-tune scored 46.2% success across 1,200 episodes on NVIDIA's RoboLab-120 benchmark, ahead of FLUX 3 Action at 42.9%, Cosmos3 Nano at 36.8% and π0.5 at 28.0%, and a fine-tune on Google's RT-1 robot data reached 80.0% on SimplerEnv against 52.4% for the original RT-1 policy; all of these figures are Runway's own and had not been independently reproduced at publication. The model runs on NVIDIA hardware from H100 and B200 data-center GPUs to RTX 5090 and RTX PRO 6000 workstations and Jetson Thor on the robot itself, with a 54.4 GB bfloat16 checkpoint and 4-bit NVFP4 variants.
Launch partners Noble Machines, Standard Bots and Ultra have been running Praxis-1 on their own hardware, on a legged humanoid, a six-axis industrial arm and a dual-arm warehouse robot, and Runway says a single checkpoint adapts to new embodiments with light fine-tuning because its video pretraining gives it a head start on how objects behave before any robot-specific data.
Praxis-1 in numbers
Values are copied from the linked source on the date shown. They are not IA Compa measurements.
- RoboLab-120 success rateas of · Source for RoboLab-120 success rate: runway.com (opens in a new tab)
- 46.2%
Where to compare Praxis-1
Independent leaderboards and official resources for comparing this model yourself. Rankings change often; each site shows its own current data.
General leaderboards
Why Praxis-1 matters for businesses
Why it matters: businesses working on warehouse, manufacturing or inspection automation can now download a frontier-class robot policy model, run it on their own hardware and fine-tune it for their own robots without a license negotiation or an API gatekeeper, which lowers the cost of piloting physical automation and keeps robot data and control inside the company.
Put it to work: AI services
Get AI News by email
Verified AI stories, each with its official source, on Monday and Thursday mornings. Add breaking news to hear about the biggest ones as soon as we publish them.
Every story links its official source
Free, in English or Spanish
Unsubscribe with one click, any time
Sources for this story
- Praxis-1 Is Now Available: Open-Weight Robotics Model (opens in a new tab)Official
Runway · runway.com
- Praxis-1 - a runwayml Collection (opens in a new tab)Official
Runway (Hugging Face) · huggingface.co
- Runway's Praxis-1 world action model for robotics is now available (opens in a new tab)Press
The Robot Report · therobotreport.com
Last verified: . IA Compa summarizes public announcements for business readers. We are not affiliated with the companies mentioned. Always confirm details, pricing, and availability with the official source.
- #runway
- #robotics
- #open-weights
- #world-models
- #apache-2-0
