# Stanford and Caltech put GPT-6 Astra in a humanoid to tidy an unfamiliar kitchen

> The Decoder reports Stanford and Caltech researchers had a humanoid robot powered by GPT-6 Astra clean an unfamiliar kitchen using HomeBody, a system where the language model calls grasping and navigation skills directly instead of relying

- **Topic**: Research
- **Published**: 2026-09-27T21:41:44.869Z
- **Canonical URL**: https://highsignal.sh/stories/stanford-and-caltech-put-gpt-6-astra-in-a-humanoid-to-tidy-an-unfamiliar-kitchen-27335bca

## Why It Matters

The project tests whether a general language model can drive a robot's low-level actions without a purpose-trained control layer. If a model can call skills like grasping and navigation directly, that changes how robot control stacks are built. The Decoder's report describes one kitchen-tidying demonstration, so it is a single result rather than proof the approach generalizes.

## Key Findings & Analysis

### What happened

The Decoder reported that researchers from Stanford and Caltech had a humanoid robot powered by GPT-6 Astra independently tidy up a kitchen it had not seen before. Their system, HomeBody, skips a specially trained control layer and lets the language model call directly into modular skills such as grasping and navigating. The account covers one demonstration; the article does not report benchmark scores, failure rates, or how often the robot succeeded.

## Primary Sources & Citations

- [Researchers plug GPT-6 Astra directly into a robot and let it clean up an unfamiliar kitchen](https://the-decoder.com/researchers-plug-gpt-6-astra-directly-into-a-robot-and-let-it-clean-up-an-unfamiliar-kitchen/) — *The Decoder* (Reporting)

---

[← Back to Headlines](https://highsignal.sh/)