Skild's S1: a robot foundation model that learns 10-min tasks from one video
rohanpaul_ai · x · 2026-08-26
Skild has released S1, a robot foundation model with in-context learning for robotics: tasks are defined by a human video demonstration rather than language instructions. Given one video showing a long multi-step task, the robot executes it immediately—no retraining, no fine-tuning, even for 10-minute tasks it has never seen.
As Rohan Paul frames it: if this scales, the enormous data bill is paid once during foundation-model training and then amortized across thousands of new tasks via prompting—potentially changing the economics of robot learning entirely.
More from Embodied
- Viral meme nails AI's irony: it can write symphonies but can't do the dishes — briannekimmel · 2026-10-16
- WEDGE turns navigation agents' own failures into targeted training worlds — mohitban47 · 2026-10-16
- Knödel Watch: animated vocabulary cards on ESP32-S3, open-sourced 4 days after unboxing — alexcovo_eth · 2026-10-16
- SlipSense fuses pressure and vibration for 23ms zero-shot robotic slip detection — micoolcho · 2026-10-16
- Stanford's Fei-Fei Li team open-sources OpenWAM, a unified world-action model framework for robots — jiqizhixin · 2026-10-16
- Robot folding leaderboard: GPT-6 Astra leads at 1349 Elo, Muse Wonder trails at 1034 — qinzytech · 2026-10-16