SkildAI releases S1: Robot model that learns tasks from a single video demo
lukas_m_ziegler · x · 2026-08-27
SkildAI released S1, a robot foundation model built around video in-context learning. By showing it a single video demonstration of an unseen task (e.g., potting plants, cooking pancakes, pour-over coffee), the model can execute it. It requires no fine-tuning or post-training, with tasks running up to 10 minutes. In a plant potting task, S1 went from demo to autonomous execution in just 33 minutes.
More from Embodied
- WEDGE uses navigation agent failures to generate targeted training environments — mohitban47 · 2026-10-16
- Tevel's robotic system harvests fruit around the clock using computer vision — LexiLove · 2026-10-16
- SHOES: 10 minutes of real-robot RL lets quadrupeds walk in shoes, CoRL 2026 paper — kscottz · 2026-10-16
- Tesla Cybercab robotaxis logged zero crashes in first two weeks of public rides, NHTSA data shows — elonmusk · 2026-10-16
- Robotics LLMs tested on 16 tasks: GPT-6 Astra hits 17.2% success vs Fable's 9.1% — pabbeel · 2026-10-16
- WEDGE paper generates training environments conditioned on navigation agent failures — _akhaliq · 2026-10-16