Runway releases a 27B open-weight model for controlling real robots

Runway released Praxis-1, a 27B-parameter mixture-of-experts diffusion transformer world action model, with open weights under the Apache 2.0 license for commercial use, including inference and fine-tuning code with full fine-tuning and LoRA support. Runway said Praxis-1 tops NVIDIA's RoboLab-120 leaderboard at 46.2% success across 1,200 episodes, versus 42.9% for FLUX 3 Action, 36.8% for Cosmos3 Nano and 28.0% for π0.5, and that a single checkpoint fine-tuned on Google's RT-1 robot reached 80.0% success on SimplerEnv against 52.4% for the original RT-1 policy. The model is available on GitHub, Hugging Face with native LeRobot support, and vLLM-Omni, with hosted inference on Reactor, and runs on NVIDIA hardware from H100 and B200 data center GPUs to RTX 5090, RTX PRO 6000 and Jetson Thor.

Praxis-1 https://runway.com/product/robotics , our state-of-the-art world action model, is now available to everyone with open weights under the Apache 2.0 license. We introduced Praxis-1 https://runway.com/research/introducing-praxis-1 in September as our first world action model, built on the same large-scale video pretraining behind our general world models. Our launch partners including Noble Machines https://www.noblemachines.ai/ , Standard Bots https://standardbots.com/ and Ultra https://www.ultra.tech/ have been running it on their own hardware: a legged humanoid, a six-axis industrial arm and a dual-arm warehouse robot. Today, the same model is open to every robotics developer and researcher. State-of-the-Art Performance Praxis-1 is a 27B-parameter mixture-of-experts diffusion transformer http://runway.com/research/praxis-1-video-pretraining-to-robot-control that predicts video and actions together. It is the top-performing world action model on NVIDIA’s RoboLab-120 leaderboard, and it outperforms every open-weight VLA. On RoboLab-120, Praxis-1 reaches 46.2% success across 1,200 episodes, compared with 42.9% for FLUX 3 Action, 36.8% for Cosmos3 Nano and 28.0% for π0.5. | Model | RoboLab-120 Success | |---|---| | Praxis-1 | 46.2% | | FLUX 3 Action | 42.9% | | Cosmos3 Nano | 36.8% | | π0.5 | 28.0% | Praxis-1 is available today on GitHub https://github.com/runwayml/praxis-1-oss , on Hugging Face https://huggingface.co/collections/runwayml/praxis-1 with native LeRobot support https://huggingface.co/docs/lerobot/main/en/praxis , and in vLLM-Omni, with hosted inference on Reactor https://reactor.inc/praxis-1 . Everything you need to run and adapt it on your own hardware is released under Apache 2.0, including for commercial use. That covers the weights and the inference and fine-tuning code, with support for both full fine-tuning and LoRA. Praxis-1 runs out of the box across NVIDIA hardware, from data center GPUs like H100 and B200, to workstation GPUs like RTX 5090 and RTX PRO 6000, to Jetson Thor on the robot itself. Adapt It to Your Hardware Praxis-1 learns how objects behave and how tasks unfold from large-scale video, which gives it a head start on a new embodiment before fine-tuning on any hardware or domain-specific data. Our research http://runway.com/research/praxis-1-video-pretraining-to-robot-control has shown that web video matches teleoperated robot video as pretraining data for policy performance, which shifts the bottleneck on a capable policy to how much general video a model can learn from. With our partners, a single Praxis-1 checkpoint has adapted to a variety of embodiments with light fine-tuning, and with LoRA you can adapt it to your own embodiment, cameras and tasks without retraining the full model. Fine-tuned on Google’s RT-1 robot, which never appeared in its training data, Praxis-1 reaches 80.0% success on SimplerEnv, well above the original RT-1 policy at 52.4%. Why Open Weight When we announced Praxis-1, we committed to shipping it as an open model. U.S. leadership in physical AI will depend on investment in both hardware and software, and on a level of interoperability and openness from American models that physical use cases don't have yet. Open weights give hardware developers the flexibility to run, inspect and adapt a world action model on their own terms. Praxis-1 builds on Runway’s work as a founding member of the Cosmos Coalition https://runway.com/news/introducing-cosmos-coalition , advancing open world models for physical AI. Together, Runway and NVIDIA are opening frontier technology stacks giving developers shared foundations to build on and solve physical AI challenges. Try it Today Download Praxis-1 on Hugging Face https://huggingface.co/collections/runwayml/praxis-1 and view the code on GitHub https://github.com/runwayml/praxis-1-oss , or try it on Runway Dev https://dev.runway.com/research/praxis-1 . Developers can post train Praxis-1 for their domain specific use cases with NeMo Automodel hands-on tutorials https://github.com/NVIDIA-NeMo/Automodel/blob/main/docs/model-coverage/wam/runwayml/praxis-1.mdx . If you're planning a deployment, contact our robotics team https://runway.com/product/robotics/contact .