Figure Introduces Helix 2.5, Tested Zero-Shot in 30 Unseen Homes

Figure released Helix 2.5, a neural network for humanoids pretrained on the Index dataset. In tests in 30 unfamiliar homes, zero-shot success rose from 9% to 56% compared with training from scratch.

• Zero-shot success rose from 9% to 56% thanks to pretraining on Index
• Tests took place in 30 Bay Area homes without on-site data collection
• The model learned cleaning, folding towels and making beds
• Figure invested $3.5 billion in compute for Helix training