Runway releases open-weight Praxis-1 robot model under Apache 2.0

1. 2:26 AM
   IThome · AINews

   Runway releases open-weight Praxis-1 robotics model trained from video

   AIRunway released Praxis-1, a robotics model that applies knowledge learned from video pretraining to real robot control, on November 11. The model is open-weight under Apache 2.0, built on a 27-billion-parameter Runway video foundation model, and can run locally, including in an NVFP4 quantized version for the RTX 5090.

1. 10:09 AM
   vLLM@vllm_projectOfficial

   vLLM adds day-0 support for Runway's Praxis-1 in vLLM-Omni

   AIvLLM says Runway's Praxis-1 is now supported in vLLM-Omni on the day of release. The integration takes camera observations, robot state, and a task instruction to generate action chunks, and supports offline inference and OpenPI / Positronic serving. vLLM says it was qualified on a single NVIDIA H100 80GB, and the model ships with open weights under Apache 2.0.

2. 10:04 AM
   Runway@runwaymlOfficial

   Runway's Praxis-1 ranks first on NVIDIA's RoboLab-120 leaderboard

   AIRunway says its Praxis-1 world action model ranks #1 on NVIDIA's RoboLab-120 leaderboard. The company says it leads every open-weight VLA on the board, including π0.5 and GR00T N1.6.

   
3. 10:04 AM
   Runway@runwaymlOfficial

   Runway makes its Praxis-1 world action model available to everyone

   AIRunway says it is making Praxis-1 available to everyone today. The company describes it as a world action model that turns video pretraining into control for real robots. Praxis-1 ships with open weights under the Apache 2.0 license, so users can run and fine-tune it on their own hardware.

   

The trend compares only continuously observed participants, so its scope may be smaller than the current heat total. Hover or tap for hourly heat; use arrow keys with a keyboard.