> ## Content Index
> Fetch the complete content index at: https://globalfeed.ai/llms.txt
> Use this file to discover other available public pages before exploring further.

# Skild AI's S1 model teaches robots new tasks from a single video
- URL: https://globalfeed.ai/en/skild-ais-s1-model-teaches-robots-new-tasks-from-a-single-video/
- Published: 2026-09-11T21:01:38.000Z
- Updated: 2026-09-11T21:01:38.000Z
- Description: Skild AI says its S1 robot foundation model, launched last week, can learn unseen tasks lasting up to 10 minutes from one video demonstration, with no retraining.
- Author: GlobalFeed Editor
- Tags: Skild AI, NVIDIA, robotics, x-nvidia, dil-en, denetim-gerekli, video

Skild AI has introduced S1, a robot foundation model built to learn previously unseen, long-horizon tasks from a single video demonstration. The model takes video as input and executes the task without updating its weights or going through task-specific post-training, an approach the company describes as in-context learning.

An operator records a video of the desired task and feeds it to the model as a prompt. S1 interprets the demonstrated intent, the objects and the sequence, then maps them into actions for the robot in front of it. According to Skild, S1 handles unfamiliar tasks lasting up to 10 minutes, including plant potting, pancake making, pour-over coffee brewing and kit assembly. In one plant-potting test, the team went from recording the demonstration to autonomous execution on hardware in 11 minutes.

## Reported results

In Skild's tests on new multistep tasks, S1 succeeded about 66% of the time at each step, against 9% for a similar AI system. Skild also estimates that one short video example can be worth roughly 380 hands-on training examples, which would take a person 50-100 hours to collect manually.

## NVIDIA stack and Foxconn deployment

S1 was built on NVIDIA AI infrastructure. Skild uses NVIDIA Cosmos world foundation models and Cosmos Curator to diversify and organize training data, Omniverse libraries and Isaac Sim for physically based virtual environments, and Isaac Lab with the Newton physics engine for reinforcement learning. The two companies are jointly developing GPU-accelerated simulation solvers for contact, grip and manipulation of solid objects, to be released to developers as part of Newton. NVIDIA Nsight tools and the TensorRT SDK cover profiling and inference optimization.

Skild, NVIDIA and Foxconn are deploying Skild Brain on dual-arm manipulators for high-precision assembly of NVIDIA Blackwell systems. In one demonstrated workflow, a robot installs a busbar and limit block, fastens 16 screws and adapts to disturbances.

The company says it reached a $100 million annual revenue run rate 10 months after its first commercial deployment and has built more than 60 deployment partnerships. Full details are on the [NVIDIA blog](https://blogs.nvidia.com/blog/skild-ai-s1-physical-ai/?ref=globalfeed.ai).