Physical Intelligence publishes pi0.5 VLA for open-world generalization
• Physical Intelligence unveiled π-0.5, an advanced vision-language-action model, on April 22, 2025. • The model was tested in real homes not included in its training data, where it successfully performed complex tasks like cleaning kitchens and bedrooms. • This demonstration marked the first instance of an end-to-end learning-enabled robotic system executing long-horizon and dexterous manipulation skills in entirely new environments.
Evidence notes
- Physical Intelligence published pi0.5 as a VLA model focused on open-world generalization across messy homes and unstructured tasks.
- The official writeup emphasizes co-training on heterogeneous data so the model can transfer knowledge across robots, instructions, and semantic task contexts.
- The release marks a clear step from lab dexterity toward general-purpose household and workplace robot behavior.
Company context
Physical Intelligence develops foundation models and learning algorithms for robots and other physically actuated systems. Its π-series models are designed to generalize across robot embodiments, tasks, and environments using robot data, language instructions, vision-language-action training, reinforcement learning, and multimodal context conditioning.