Skild AI released S1, a robotics foundation model that executes tasks up to 10 minutes long from a single human video prompt with no fine-tuning. The company reports 66% success on unseen tasks versus 9% for language-prompted VLAs at the same 100k-hour training scale, with demonstrations covering pancake flipping, pour-over coffee, plant potting, and kit assembly.
Sequoia's Alfred Lin called single-prompt execution of long-horizon tasks 'a game changer.' The model represents a significant step toward general-purpose robotics, where robots can learn complex workflows from minimal human demonstration rather than requiring extensive programming or task-specific training.
This breakthrough addresses one of robotics' longstanding challenges: scaling task learning without requiring massive datasets for each individual skill. By learning from video demonstrations in context, S1 could dramatically reduce the cost and time needed to deploy robots in diverse real-world settings.
