# Skild S1 机器人基础模型：视频即提示词

- 来源：Rohan Paul (@rohanpaul_ai)
- 发布时间：2026-08-26 02:05
- AIHOT 分数：44
- AIHOT 链接：https://aihot.virxact.com/items/cmt8znak605k3rolylr1cd012
- 原文链接：https://x.com/rohanpaul_ai/status/2092312459382235407

## AI 摘要

Skild 发布机器人基础模型 S1，用视频演示替代语言指令来定义任务。只需提供 1 段人类视频，S1 即可直接执行长达 10 分钟的多步骤新任务，无需重新训练或微调。若规模化，基础模型训练的高昂数据成本可摊薄至数千个新任务，有望改变机器人学习的经济性。

## 正文

Robotics is more and more getting close to get its version of in-context learning.

Skild just released S1, a robot foundation model that uses video demonstrations to define tasks instead of language instructions.

Give S1 1 human video showing a long, multi-step task, and the robot executes it straight away. No retraining. No fine-tuning.

If this scales, you pay the enormous data bill once during foundation-model training, then amortize it across thousands of new tasks through prompting. That could change the economics of robot learning completely.

### 引用推文

> Skild AI：Introducing S1, our new foundation model that learns from one example. It can be taught 10-minute long tasks that it has never seen before, from one video promp...