Runway Releases Praxis-1, a 27-Billion-Parameter Open-Weight Robot Action Model

Runway has released Praxis-1, a 27-billion-parameter world action model for robotics, under an Apache 2.0 license that outperforms Nvidia's Cosmos and Google's RT-1 on published benchmarks.

Key Takeaways

  • Runway released Praxis-1, a 27-billion-parameter robot action model, under an Apache 2.0 open-weight license on November 22, 2026.
  • Praxis-1 scored a 46.2% success rate on the RoboLab-120 benchmark, ahead of Nvidia's Cosmos 3 Nano (36.8%) and Physical Intelligence's π0.5 (28%).
  • After fine-tuning for Google's RT-1 robot, Praxis-1 reached 80% success on SimplerEnv, up from RT-1's own 52.4% baseline.
  • The model combines video prediction with robot action generation, letting it learn from general video rather than only robot-specific data.
  • Runway demonstrated the model on industrial arms, warehouse robots and humanoids with partners Noble Machines, Standard Bots and Ultra.

Runway Releases Praxis-1, a 27-Billion-Parameter Open-Weight Robot Action Model

Runway made its Praxis-1 robot foundation model publicly available on November 22, 2026, releasing the 27-billion-parameter model's weights, inference software and fine-tuning tools under an Apache 2.0 license. The model, first introduced in September 2026, combines video prediction with robot action generation so that systems can learn physical tasks from general video footage rather than relying solely on robot-specific demonstration data, according to Robotics and Automation News.

On the RoboLab-120 benchmark, Praxis-1 posted a 46.2% success rate, ahead of FLUX 3 Action (42.9%), Nvidia's Cosmos 3 Nano (36.8%) and Physical Intelligence's π0.5 (28%). After fine-tuning for Google's RT-1 robot on the SimplerEnv benchmark, Praxis-1 reached an 80% success rate, well above RT-1's own baseline of 52.4%.

Why an Open Weight Release Matters Here

Most comparably sized robot foundation models have shipped as closed, API-gated products, including Nvidia's Cosmos line. Runway's decision to publish full weights on GitHub and Hugging Face, with support for the LeRobot framework, hands smaller robotics teams a model they can fine-tune on their own hardware without paying per-call API fees or sending proprietary task data to a third party.

That matters because the practical bottleneck in embodied AI right now isn’t raw model capability, it's getting any one model to generalize across the dozens of arm, gripper and mobile-base configurations startups actually use. An open-weight release lets a hardware maker adapt Praxis-1 directly to its own robot rather than waiting for a vendor to add support, which is exactly how Runway demonstrated the model, in partnership with Noble Machines, Standard Bots and Ultra across industrial arms, dual-arm warehouse robots and humanoid platforms.

The Benchmark Gap Versus the Deployment Gap

The RoboLab-120 numbers put Praxis-1 roughly 3 to 10 percentage points ahead of its closest named rivals, a real but incremental lead rather than a decisive one. Runway itself has flagged the limits plainly, acknowledging the model "still struggles with visually similar objects, ambiguous instructions and complex tasks requiring multiple sequential actions," which covers a large share of real warehouse and factory picking scenarios.

The more consequential number may be the SimplerEnv jump, from 52.4% to 80% after fine-tuning on RT-1 specifically. That suggests Praxis-1's video-pretrained foundation transfers well to new hardware with comparatively modest additional training, which is the quality that determines whether a foundation model actually displaces bespoke, single-robot policies. Nvidia, Physical Intelligence and Google's own RT-X lineage will all be judged against that same transfer-efficiency bar going forward, not just raw success rates on a fixed benchmark.

Category: Machine Learning

Tags: embodied AI robot foundation model Runway Praxis-1 Nvidia Cosmos open weight AI

Related Articles

Frequently Asked Questions

What is Runway's Praxis-1?

A 27-billion-parameter open-weight AI model that generates robot actions by combining video prediction with robotic control, released under an Apache 2.0 license on November 22, 2026.

How does Praxis-1 compare to Nvidia's Cosmos model?

Praxis-1 scored 46.2% on the RoboLab-120 benchmark versus 36.8% for Nvidia's Cosmos 3 Nano.

Can developers fine-tune Praxis-1 for their own robots?

Yes, Runway released weights, inference software and fine-tuning tools via GitHub and Hugging Face with LeRobot framework support.

What are Praxis-1's known limitations?

Runway says the model struggles with visually similar objects, ambiguous instructions, and complex multi-step sequential tasks.