A VLA with Open-World Generalization
Our latest generalist policy, π0.5, extends π0 and enables open-world generalization. Our new model can control a mobile manipulator to clean up an entirely new kitchen or bedroom.
A VLA with Open-World Generalization π 0.5 : a VLA with Open-World Generalization Published April 22, 2025 Email research@physicalintelligence.company Kevin Black, Noah Brown, James Darpinian, Karan Dhabalia, Danny Driess, Adnan Esmail, Michael Equi, Chelsea Finn, Niccolo Fusai, Manuel Galliker, Dibya Ghosh, Lachy Groom, Karol Hausman, Brian Ichter, Szymon Jakubczak, Tim Jones, Liyiming Ke, Devin LeBlanc, Sergey Levine, Adrian Li-Bell, Mohith Mothukuri, Suraj Nair, Karl Pertsch, Allen Ren, Lucy Xiaoyang Shi, Laura Smith, Jost Tobias Springenberg, Kyle Stachowicz, James Tanner, Quan Vuong, Home
saved by
related reading
- A VLA with Open-World Generalizationphysicalintelligence.company
- A Steerable Model with Emergent Capabilitiespi.website
- Physical Intelligence (π)pi.website
- How Claude Performs on Robotics Tasks \ Anthropicanthropic.com
- Towards Machines with a Thousand Hands - Generalist AIgeneralistai.com
- Physical Intelligence (π)physicalintelligence.company
- State of Vision-Language-Action (VLA) Research at ICLR 2026 – Moritz Reussmbreuss.github.io
- $π_0$: A Vision-Language-Action Flow Model for General Robot Controlalphaxiv.org
- State of Robot Learning, December 2025vedder.io
- Emergence of Human to Robot Transfer in Vision-Language-Action Modelspi.website
- SimpleVLA-RL: Scaling VLA Training via Reinforcement Learningalphaxiv.org
- Pretrained to Imagine, Fine-Tuned to Act: The Rise of World-Action Models | NVIDIA Technical Blogdeveloper.nvidia.com