τ0-VLA: Hierarchical Robot Foundation Model with World-Model-Guided Test-Time Search

sii-research · hf · 2026-08-22

τ0-VLA is a hierarchical vision-language-action (VLA) robot foundation model. Its key idea is applying world-model-guided test-time computation to high-level subtask decisions.

Released on Hugging Face under sii-research.

Original post →

More from Embodied

Embodied channel →