NVIDIA blog: Skild AI’s S1 learns long robot tasks from one video

10 Sep 2026: NVIDIA published that Skild AI’s S1 robot foundation model can learn previously unseen long-horizon tasks from a single video demonstration via in-context learning — no weight update. Company blog is the primary. Performance and $100M ARR figures are Skild’s / NVIDIA’s reporting.

10 Sep 2026: NVIDIA published “Skild AI Taps NVIDIA Physical AI to Teach Robots New Tasks From a Single Video.” S1 is designed to learn previously unseen long-horizon tasks from a single video demonstration. The model interprets demonstrated intent, objects, and sequence, then maps them to actions without updating weights or running task-specific post-training — in-context learning. That dated NVIDIA blog is the filing event.

NVIDIA/Skild: S1 can perform unfamiliar tasks lasting up to about 10 minutes — plant potting, pancake making, pour-over coffee brewing, kit assembly — spanning dozens of manipulation steps and skill sequences the model had not previously performed.

Skild tests, as reported on NVIDIA’s page: on new multistep tasks, S1 succeeded about 66% of the time at each step versus 9% for a similar AI system — more than sevenfold. Skild estimates one short video can be as useful as roughly 380 hands-on training examples, which a person collecting manually could take 50–100 hours. Those figures are Skild’s.

NVIDIA: Skild built S1 on NVIDIA AI infrastructure. The collaboration spans synthetic data (Cosmos), Isaac Lab / Isaac Sim, Omniverse, and TensorRT. Newton physics and joint GPU-accelerated simulation solvers are mentioned as joint work, with solvers to be made available in Newton.

Deepak Pathak, Skild CEO, on NVIDIA’s page: “Learning by experience, and not preprogramming, is the step change that has happened in robotics.” He cited Isaac Lab and Cosmos for scalable, diverse experience across scenarios and embodiments.

Commercial framing on the same post: Skild reached a $100 million annual revenue run rate 10 months after its first commercial deployment, and has more than 60 deployment partnerships spanning manufacturing, logistics, inspection, security, food preparation, and other applications. Treat that as Skild’s claim via NVIDIA, not audited financials.

Factory demo: Skild, NVIDIA, and Foxconn are deploying Skild Brain on dual-arm manipulators for high-precision assembly of NVIDIA Blackwell systems. NVIDIA’s description of a demonstrated workflow: a robot installs a busbar and limit block, fastens 16 screws, and adapts to disturbances across a multistep task.

A dated NVIDIA primary documents a commercial physical-AI path where a robot foundation model takes a single operator video as the prompt for multistep factory tasks — with named NVIDIA stack pieces and a Foxconn/Blackwell assembly example. Capability and revenue numbers stay the companies’ claims.