Figure's Helix 2.5 achieves zero-shot whole-body generalization across 30 unseen homes
coreylynch · x · 2026-09-18
Figure unveiled Helix 2.5, its most advanced neural network, pretrained on Index, its global-scale human-behavior dataset. Highlights:
- Zero-shot whole-body autonomy in 30 real Bay Area homes: tidying living rooms, folding towels, and making beds with no data collection, fine-tuning, or adaptation — billed as the first demonstration of zero-shot whole-body generalization at this scope on a humanoid.
- One foundation model, three whole-body behaviors spanning locomotion, rigid and deformable manipulation, bimanual coordination, and active perception.
- Index pretraining drives generalization: with everything else fixed, zero-shot success jumps from 8% to 56% (alternatively 9%→56%, 6x+).
- Predictable scaling: doubling Index data let the team forecast the largest run's loss to four decimal places before training.
Figure has committed $3.5B of compute for the next Helix version.
Related event: Figure Unveils Helix 2.5: Zero-Shot Household Work Across 30 Unseen Homes(27 posts)→
More from Embodied
- Autonomous's Harness Pro uses an ESP32-P4, not a mini PC, to monitor AI coding agents — dee_hw · 2026-11-02
- KAIST's QCP lets robot policies control execution quality in real time — kaist-ai · 2026-11-02
- ANYbotics begins customer deliveries of ANYmal X, first legged robot certified for explosive atmospheres — lukas_m_ziegler · 2026-11-02
- InSpatio opens China's first embodied simulation data center targeting 1M hours/month — 智东西 · 2026-11-02
- GR-BC mines better trajectories from imperfect demos, hitting ~91% success on precise manipulation — 机器之心 · 2026-11-02
- RAM prices tripled, so Tesla cut a third of memory from Optimus's brain — aakashgupta · 2026-11-02