GEN-1.5 embodied model learns tasks from single demos, corrects errors
SongShuran · x · 2026-08-20
GeneralistAI released GEN-1.5, their latest embodied foundation model. It can learn new tasks prompted with just 3-12 seconds of a single human demonstration, requiring no gradient updates or fine-tuning. The model generalizes prompts to new situations, recovers from mistakes, and improvises new strategies to reach the same goal.
Related event: Generalist AI Unveils GEN-1.5: One-Shot Robot Learning in Seconds(10 posts)→
More from Embodied
- Zetta ζ: Closed-Loop Embodied Harness Evolves Critics and Recovery Skills Online — Xin Ding · 2026-08-20
- SoftVTBench: A Deformation-Aware Visuo-Tactile Dataset for Deformable-Object Manipulation — TuojingAI · 2026-08-20
- Robotics manipulation harder than pure vision due to problem definition — chris_j_paxton · 2026-08-20
- Zipline on drone noise mitigation and preventing delivery screw-ups — kscottz · 2026-08-20
- Tesla aims for Optimus commercial sales in 2H27; FSD V15 hailed as major leap — chrisfirst · 2026-08-20
- Beijing robot output up 2.3x in H1; Zhiyuan A3 wins China & EU certifications — 创业邦 · 2026-08-20