Skild AI unveils S1: robot foundation model learns 10-minute novel tasks from a single video prompt
deepakpathak · x · 2026-08-26
Skild AI introduced S1, a new foundation model built around in-context learning:
- Learns tasks it never saw in pre-training from a single video prompt, with no fine-tuning;
- Handles long-horizon tasks over 10 minutes;
- Real-time operation demos were released.
Researcher Deepak Pathak framed it as building intelligence "from the foundations up, not from the top down," declaring that in-context learning for robotics has arrived.
Related event: Skild's S1 Learns New Robot Tasks from a Single Video Demo(3 posts)→
More from Embodied
- Robotics breakthrough accelerates timeline to generalized humanoids: Scobleizer — HankYeomans · 2026-08-26
- Skild robot learns to make pancakes from one video, demonstrating long-horizon task execution — Scobleizer · 2026-08-26
- Late 2020s will begin a proper sci-fi era with AGI, robots, and TERA factories — Dr_Singularity · 2026-08-26
- M6 Mac Mini starts at $899 — andrejusb · 2026-08-26
- Scobleizer: Skilled's In-House Data Strategy May Force Industry Rethink — Scobleizer · 2026-08-26
- Tesla Cybertaxi with Texas plates spotted in San Diego — justin_hart · 2026-08-26