# Nvidia's Astranomy study reaches 74.8% success in robot control with GPT-6 Astra

Digest AI · Research · published 2026-10-20T14:37:00Z

Canonical: https://digestai.news/story/nvidia-s-astranomy-study-reaches-74-8-success-in-robot-control-with-gp

## Summary

In October 2026, NVIDIA's Seattle Robotics Lab released Astranomy, a study that tests whether OpenAI's GPT‑6 Astra can directly control robots without extra training. The system feeds camera images, robot state, and task instructions to the model and generates actions via API calls. Using the RoboLab‑120 benchmark of 120 manipulation tasks in the DROID simulation, Astranomy achieved a 74.8% success rate, surpassing the best specialized robot model, Lakeside, which scored 51.0%.

The research compared three control approaches. The Agentic Policy, which runs GPT‑6 Astra inside a persistent Codex session, reached the highest success at 74.8%, using about 30.9 million tokens per successful episode and costing an estimated $36.40. Direct Policy, which calls the model without history, scored 66.3% and required 18.56 seconds per inference. Orchestrated Policy, which supervises NVIDIA’s Cosmos3‑Nano‑Policy, achieved 66.5%, showing little gain from combining a specialized model. Motion‑representation tests showed EEF methods outperform joint‑position and B‑spline formats, especially under the Agentic setup.

Authors note the results are limited to simulation; real‑world latency and API costs could hinder high‑speed operation, and performance on assembly or rapid‑change tasks remains untested.

## Key points

- Astranomy achieved 74.8% success on the RoboLab-120 benchmark using GPT‑6 Astra, beating the best specialized model by 23.8 points.
- Agentic Policy with Codex used about 30.9 million tokens per successful episode, costing an estimated $36.40.
- Direct Policy inference took 18.56 seconds per call; simulation latency does not reflect real‑world speed constraints.

## Why it matters

Demonstrates that general‑purpose multimodal AI can control robots without task‑specific training, potentially lowering development costs and accelerating deployment of AI‑driven automation.

## Sources

1. [Overview of NVIDIA 'Astranomy' - Achieving a 74.8% Success Rate in Robot Control with GPT-6 Astra](https://note.com/npaka/n/n3c80ede3bf24?hl=en) (note.com, 2026-10-20)

Part of the developing story: [AI Model Showdown Across Platforms](https://digestai.news/thread/replit-ends-partnership-with-anthropic-cites-cost-and-performance) (3 stories)

## Cite

Digest AI, "Nvidia's Astranomy study reaches 74.8% success in robot control with GPT-6 Astra", 20 October 2026, https://digestai.news/story/nvidia-s-astranomy-study-reaches-74-8-success-in-robot-control-with-gp

---

Written by Digest AI's editorial model from the linked sources; the sources are the record. Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse
JSON: https://digestai.news/story/nvidia-s-astranomy-study-reaches-74-8-success-in-robot-control-with-gp.json