---
format: "aidr-story-markdown/v1"
id: "3be14d46c4a7d95b99c833879c054069072226dcb7536b357649ff8e85a24266"
canonical_url: "https://aidr.today/3be14d46?lang=en"
title: "GPT-6 Astra Pro Renders 100.1 Million Blocks in First Model to Handle All MineBench Prompts"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-09-08T05:41:12.000Z"
category: "Models"
topics: ["openai","gpt","multimodal","benchmark","reasoning","llm","research"]
source_urls: ["https://huggingnews.com/ai/gpt-6-astra-pro-renders-1001-million-blocks-in-first-model-to-handle-all-100fc656","https://x.com/minebench_ai/status/2097052533961248887","https://x.com/BorisMPower/status/2097075023756587131","https://x.com/nicdunz/status/2097101602649739751","https://x.com/nicdunz/status/2097102538088542410","https://x.com/scaling01/status/2097032985019023606","https://x.com/scaling01/status/2097139066370207868","https://x.com/htihle/status/2096982137412931850"]
summary: "OpenAI's GPT-6 Astra Pro produced a programmatic representation of New York City spanning 8.2 kilometers edge to edge. The model operated on an 8,192 cubed grid to generate a 1.65 gigabyte build expanded into JSON. This specific output utilized 100.1 million blocks and originated from a program sized at approximately 61 kilobytes. The output represents a 32,768 fold increase in possible voxel positions over the 256 cubed grid standard on the MineBench testing platform. MineBench described Astra Pro as the first model in its testing to handle any prompt provided. The project used a scale of one meter per block to create a dense and accurate cityscape."
---

# GPT\-6 Astra Pro Renders 100\.1 Million Blocks in First Model to Handle All MineBench Prompts

> [Open the canonical story](<https://aidr.today/3be14d46?lang=en>)

**Published:** 2026-09-08T05:41:12.000Z
**Category:** Models
**Topics:** openai, gpt, multimodal, benchmark, reasoning, llm, research

## Summary

OpenAI's GPT\-6 Astra Pro produced a programmatic representation of New York City spanning 8\.2 kilometers edge to edge\. The model operated on an 8,192 cubed grid to generate a 1\.65 gigabyte build expanded into JSON\. This specific output utilized 100\.1 million blocks and originated from a program sized at approximately 61 kilobytes\. The output represents a 32,768 fold increase in possible voxel positions over the 256 cubed grid standard on the MineBench testing platform\. MineBench described Astra Pro as the first model in its testing to handle any prompt provided\. The project used a scale of one meter per block to create a dense and accurate cityscape\.

## Sources

- [Story source](<https://huggingnews.com/ai/gpt-6-astra-pro-renders-1001-million-blocks-in-first-model-to-handle-all-100fc656>)
- [Story source](<https://x.com/minebench_ai/status/2097052533961248887>)
- [Supporting source](<https://x.com/BorisMPower/status/2097075023756587131>)
- [Story source](<https://x.com/nicdunz/status/2097101602649739751>)
- [Supporting source](<https://x.com/nicdunz/status/2097102538088542410>)
- [Supporting source](<https://x.com/scaling01/status/2097032985019023606>)
- [Supporting source](<https://x.com/scaling01/status/2097139066370207868>)
- [Story source](<https://x.com/htihle/status/2096982137412931850>)