---
format: "aidr-story-markdown/v1"
id: "a0d9a6f04b030ac0afb7bdd548f55997e469475f419ba3bc64704b94a07efc3f"
canonical_url: "https://aidr.today/a0d9a6f0?lang=en"
title: "GPT-6 Astra Beats Standard Factorio Game With Enemies Enabled for $4,500"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-09-12T08:40:12.000Z"
category: "Agents"
topics: ["openai","gpt","agent","reasoning","benchmark"]
source_urls: ["https://huggingnews.com/ai/gpt-6-astra-beats-standard-factorio-game-with-enemies-enabled-for-4500-f3beb29b","https://x.com/mattparlmer/status/2098655487453757863","https://x.com/steipete/status/2098616501142208831","https://x.com/chris_j_paxton/status/2098565917575647338"]
summary: "The GPT-6 Astra model overcame the automation game Factorio with enemies active after nearly 44 hours of simulation time. The run required 4 days 11 hours of total processing on a Codex goal clock and cost roughly $4,500 in API fees. Costs could decrease 50 times if the system offloads tasks to Luna or utilizes standard scripts for common operations. The system failed to solve embodied tasks, though reports suggest the result indicates robotics may eventually adopt similar scaling laws. Future benchmarks include using Astra to coach other AI models, such as Muse Spark or Qwen 3.8 27B, to solve random seed challenges."
---

# GPT\-6 Astra Beats Standard Factorio Game With Enemies Enabled for $4,500

> [Open the canonical story](<https://aidr.today/a0d9a6f0?lang=en>)

**Published:** 2026-09-12T08:40:12.000Z
**Category:** Agents
**Topics:** openai, gpt, agent, reasoning, benchmark

## Summary

The GPT\-6 Astra model overcame the automation game Factorio with enemies active after nearly 44 hours of simulation time\. The run required 4 days 11 hours of total processing on a Codex goal clock and cost roughly $4,500 in API fees\. Costs could decrease 50 times if the system offloads tasks to Luna or utilizes standard scripts for common operations\. The system failed to solve embodied tasks, though reports suggest the result indicates robotics may eventually adopt similar scaling laws\. Future benchmarks include using Astra to coach other AI models, such as Muse Spark or Qwen 3\.8 27B, to solve random seed challenges\.

## Sources

- [Story source](<https://huggingnews.com/ai/gpt-6-astra-beats-standard-factorio-game-with-enemies-enabled-for-4500-f3beb29b>)
- [Story source](<https://x.com/mattparlmer/status/2098655487453757863>)
- [Supporting source](<https://x.com/steipete/status/2098616501142208831>)
- [Supporting source](<https://x.com/chris_j_paxton/status/2098565917575647338>)