SkildAI Unveils S1: Robotic Model Learning from Single Video Demo

deepakpathak · x · 2026-08-27

SkildAI introduced S1, a foundation model designed to learn from one example. S1 can perform 10-minute long, unseen robotic tasks from just a single video prompt without fine-tuning, operating in real-time via in-context learning.

Related event: Skild Unveils S1: Robots Learn 10-Minute Tasks from a Single Video(33 posts)→

Original post →

More from Embodied

Embodied channel →