SkildAI Unveils S1: Robot Model Learns Tasks from One Video
deepakpathak · x · 2026-08-26
SkildAI has released S1, a new robotic foundation model with powerful in-context learning capabilities.
- Core Capability: Can learn complex tasks up to 10 minutes long from a single video prompt and execute them in real-time in the wild without any fine-tuning. This demonstrates strong generalization capabilities.
- Significance: Dubbed a key unlock via "video prompting," it promises to enable robots to perform unseen tasks in unstructured environments. Deepak Pathak praised the team's vision.
More from Embodied
- Tesla upgrades in-car Grok to world's #1 speech-to-speech AI — XFreeze · 2026-08-26
- Deepak Pathak Demonstrates In-Context Learning for Robotics with One-Shot Video Prompts — deepakpathak · 2026-08-26
- Seeking AI tools for generating 3D printable minis with watertight meshes — Minus8058 · 2026-08-26
- Cars and Humanoid Robots May Be Closer Industries Than Expected — VraserX · 2026-08-26
- Sergey Levine on humanoid robotics today: China vs US and landing timelines — svlevine · 2026-08-26
- SkildAI Founder: S1 Demonstrates Physical Steerability in Long-Horizon Tasks — deepakpathak · 2026-08-26