Figure's Helix 2.5 achieves zero-shot whole-body autonomy across 30 unseen homes
FinanceYF5 · x · 2026-09-22
Figure has unveiled Helix 2.5, its most advanced neural network to date, built to answer one question: can a humanoid enter a home it has never seen and immediately get to work?
Key points:
- Zero-shot across 30 real homes: The robot performed three long-horizon behaviors — tidying living rooms, folding towels, and making beds — in 30 Bay Area homes with zero data collection, fine-tuning, or adaptation. Figure claims this is the first demonstration of zero-shot whole-body generalization at this scope on a humanoid.
- One foundation model, three behaviors: A single base model pretrained on Index, Figure's global-scale human behavior dataset, covers locomotion, rigid and deformable manipulation, bimanual coordination, and active perception.
- Index pretraining drives generalization: With task data, architecture, and training held fixed, Index pretraining was the key differentiator.
- Figure also released a 4-hour full test video showing real performance rather than cherry-picked successes.
Related event: Figure's Helix 2.5 Does 56% of Chores Zero-Shot in 30 Unfamiliar Homes(4 posts)→
More from Embodied
- Designing clothes for humanoid robots: fabric serves the mechanism, not the body — yongqianme · 2026-09-22
- Inside robot data collection: First Contact adds mocap to its setup — Scobleizer · 2026-09-22
- Robot arm grasps cube 71 minutes in: model writes its own kinematics, no training — Scobleizer · 2026-09-22
- NUS's Grounded Action Model Tops Robot Manipulation Benchmarks with 3D Grounding — NationalUniversityofSingapore · 2026-09-22
- Distilling World-Model Features into VLAs: 0.8B Policy Hits 97.9% on LIBERO — Trung Dao · 2026-09-22
- HIRO Industries Unveils Origin: A 17-DOF Dual-Arm Robot for Packing and Kitting Workstations — Scobleizer · 2026-09-22