GPT-5.6 Sol can beat Factorio on peaceful mode
What happened
GPT-5.6 Sol, running in Codex with a low thinking level, autonomously completed a peaceful Factorio run in 13.5 hours of game time, reaching the rocket launch despite needing manual effort to fix late‑game production connections. The run was monitored with a game manager that let the model interact programmatically while respecting normal player restrictions.
Why it matters
The success shows that autonomous gameplay in Factorio can serve as a clear, observable testbed for evaluating AI model capabilities and their improvement over earlier versions like GPT‑5.5.
Is this a big deal?
Get stories like this every morning
The day's AI news in plain English, in your inbox at 8:30 IST. Free.
More from OpenAI
Introducing ChatGPT WalletOpenAI · 13 Nov 2026Shield helps technicians hand off legacy portal work with OpenAIOpenAI · 13 Nov 2026OpenAI launches ChatGPT Wallet, AI agents pay for purchaseOpenAI · 13 Nov 2026OpenAI disbands its robotics team as its general models start running robotsOpenAI · 13 Nov 2026Yang-Mills, OpenAI proof, and the mass gap problemOpenAI · 13 Nov 2026This page was collected from ryanmadden.net and summarised automatically by AIPulsee's pipeline. See how it runs.