--- title: GPT-5.6 Sol can beat Factorio on peaceful mode | daily.dev description: GPT-5.6 Sol, running in Codex at low thinking level, completed a peaceful Factorio run (no biters) in 13.5 hours of game time. The agent interacted with the... canonical: https://daily.dev/posts/gpt-5-6-sol-can-beat-factorio-on-peaceful-mode-pwbl65ste twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: GPT-5.6 Sol can beat Factorio on peaceful mode | daily.dev og:description: GPT-5.6 Sol, running in Codex at low thinking level, completed a peaceful Factorio run (no biters) in 13.5 hours of game time. The agent interacted with the... og:url: https://daily.dev/posts/gpt-5-6-sol-can-beat-factorio-on-peaceful-mode-pwbl65ste og:image: https://api.daily.dev/og/posts/PWbl65STe.png og:image:alt: GPT-5.6 Sol can beat Factorio on peaceful mode og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # GPT-5.6 Sol can beat Factorio on peaceful mode **[Hacker News](https://daily.dev/sources/hn)** · 2 min read · 1 upvotes · 0 comments ## Summary GPT-5.6 Sol, running in Codex at low thinking level, completed a peaceful Factorio run (no biters) in 13.5 hours of game time. The agent interacted with the game through a Python library that enforced the same reach and fog-of-war limits human players face. Scaling issues appeared later than expected, slowing progress. The rocket launch was only achieved through significant manual effort, as Sol ferried supplies between areas to paper over failed production connections. This is an improvement over GPT-5.5, which suffered from failures to use transport belts at all. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### Can GPT-5.6 complete a full Factorio run autonomously, including a rocket launch? Yes, GPT-5.6 Sol running in Codex at low thinking level completed a peaceful Factorio run and launched a rocket after 13.5 hours of game time. It used a Python library enforcing human-like restrictions such as player reach and fog of war. The rocket launch required Sol to manually ferry supplies between areas to compensate for failed late-game production connections. _Developers tracking how agentic coding models handle long, complex tasks can follow these capability tests on daily.dev._ ### How does GPT-5.6's game-playing ability compare to GPT-5.5 in Factorio? GPT-5.6 Sol showed a clear improvement over GPT-5.5 in Factorio, successfully using transport belts and completing a full peaceful run through rocket launch. GPT-5.5, by contrast, failed to use transport belts at all during earlier attempts. Scaling issues still appeared for GPT-5.6, but later in the run than expected, during the post-blue-science production stage. _Anyone comparing successive model releases for agentic reliability can keep up with these benchmarks on daily.dev._ ## Community take How the wider developer community reacted, aggregated from 1 discussion and 133 comments across hackernews (as of 2026-10-31). **TL;DR:** Discussion is split between people fascinated by the Factorio run's implications for AI-in-games and skeptics noting the agent leaned heavily on manual player-carrying instead of true belt automation, with extensive debate on whether training data and peaceful mode make the feat less impressive. **Sentiment:** 30% positive · 45% mixed · 25% skeptical **The case for** - Some see it as proof LLMs can produce bespoke (if odd) factory designs rather than just copying examples. - A few note the run is a clear improvement over the prior model that couldn't use belts at all. - Several find it a more honest/interesting benchmark than other flashy agent demos like vending-machine simulators. **The pushback** - Multiple commenters point out the agent largely avoided belts and instead manually ferried items via the player character, missing the core automation mechanic. - Several argue peaceful mode removes real-time pressure and biters, making the challenge mostly a spreadsheet/logistics puzzle rather than a true test. - Many suspect the model's performance is inflated by having ingested the entire Factorio wiki, blueprints, and tutorial videos, rather than genuine problem-solving from scratch. - Some note the lack of time pressure or fatigue lets the AI get away with inefficient manual-labor strategies a human wouldn't tolerate. **By community** - hackernews (mixed): Conversation splits between enthusiasm about AI agents in games/future NPC applications and skepticism about how much the Factorio result is undercut by peaceful mode, training data, and reliance on manual item-carrying instead of belts. **Hottest debate:** Whether the achievement is genuinely impressive or largely an artifact of removing time pressure (peaceful mode) and having memorized exhaustive Factorio wiki/blueprint content. **Open questions** - How would the agent handle the added complexity and real-time pressure of biters/combat? - Would it perform as well on a game not extensively documented online, without memorized guides and blueprints? - What exactly is the underlying harness/integration layer, and will it support blueprints or expansions like Space Age? **Highlights** > It doesn't seem to understand the core mechanic of Factorio, which is putting outputs onto belts and routing them to other inputs. It does this in a few places, but for the most part everything it put into chests and moved around by the player. > — [fasterik on hackernews · 5 comments](https://news.ycombinator.com/item?id=50319387) > One limitation of these type of demos (and similar with Minecraft, etc) is that: 1. Putting into "peaceful" mode makes response time a non-factor 2. These models have read and have access to every guide and video ever published on the subject. Most humans, especially on first play, learn game systems and interactions from scratch, then develop novel strategies and plans and "hypothesis test" their way into beating the game. This is still impressive compared to where we were a few years ago, and being able to synthesize all existing information about a game and then play it to completion in a semi-realistic setting is great. At a quick glance someone might mistake this for "brilliant AI solves Factorio", when in reality, given a similar but different game where it wasn't gifted all the information or training data (e.g. massive videos and screenshots) it might fare terribly (as others have pointed out in comments). > — [program\_whiz on hackernews · 3 comments](https://news.ycombinator.com/item?id=50319877) > It is weird and almost alien how AI designs factories. When Claude announced Fable 5 they also did a Factorio demo that was the same kind of designs and a lot of hand feeding and direct insertion. In some sense it proves to me that llms are not just copying and are actually creating a bespoke solution. But I honestly expected more from frontier models > — [kanemcgrath on hackernews · 3 comments](https://news.ycombinator.com/item?id=50319355) > Back of the envelope math: If you measure by the product of throughput and distance traveled, for an item with a stack size of 100 (e.g. metal plates), the player's inventory adds up to roughly the same capacity as 2500 yellow belts. And that's assuming 100% utilization running back and forth in both directions, with no unused inventory space and no wasted time. A basic assembler can manufacture 2500 belts in about 21 minutes. So as a rule of thumb, if you're playing for longer than 21 minutes, belts are worth the investment (at least for bulk transport of raw materials). (I guess maybe add an additional factor of at most √2, since the player can run diagonally and belts are orthogonal.) > — [teraflop on hackernews](https://news.ycombinator.com/item?id=50320016) > (Author) I did this on the 'Plus' plan over a series of 5h quota segments so $20 for me, but at API pricing the tokens would have been $52 > — [ryanmadden on hackernews](https://news.ycombinator.com/item?id=50319825) **Source threads** - [hackernews](https://news.ycombinator.com/item?id=50318871) · 93 points · 133 comments ## Similar posts on daily.dev - [Why GPT-5.6 Sol is a Liability for Code](https://daily.dev/posts/why-gpt-5-6-sol-is-a-liability-for-code-smitzabev) · Medium · 0 upvotes · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#llm](https://daily.dev/tags/llm), [#gpt](https://daily.dev/tags/gpt), [#openai-codex](https://daily.dev/tags/openai-codex) [View this post on daily.dev](https://daily.dev/posts/gpt-5-6-sol-can-beat-factorio-on-peaceful-mode-pwbl65ste) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"GPT-5.6 Sol can beat Factorio on peaceful mode","url":"https://daily.dev/posts/gpt-5-6-sol-can-beat-factorio-on-peaceful-mode-pwbl65ste","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/gpt-5-6-sol-can-beat-factorio-on-peaceful-mode-pwbl65ste"},"datePublished":"2026-10-31T19:50:06.606Z","dateModified":"2026-10-31T21:21:10.488Z","description":"GPT-5.6 Sol, running in Codex at low thinking level, completed a peaceful Factorio run (no biters) in 13.5 hours of game time. The agent interacted with the...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/fe6c227110883e4febe9fa818fd4a8cc?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/fe6c227110883e4febe9fa818fd4a8cc?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Hacker News","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Hacker News","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/hn","url":"https://daily.dev/sources/hn"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/gpt-5-6-sol-can-beat-factorio-on-peaceful-mode-pwbl65ste","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,llm,gpt,openai-codex","timeRequired":"PT2M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Hacker News","item":"https://daily.dev/sources/hn"},{"@type":"ListItem","position":3,"name":"GPT-5.6 Sol can beat Factorio on peaceful mode"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/gpt-5-6-sol-can-beat-factorio-on-peaceful-mode-pwbl65ste#faq","mainEntity":[{"@type":"Question","name":"Can GPT-5.6 complete a full Factorio run autonomously, including a rocket launch?","acceptedAnswer":{"@type":"Answer","text":"Yes, GPT-5.6 Sol running in Codex at low thinking level completed a peaceful Factorio run and launched a rocket after 13.5 hours of game time. It used a Python library enforcing human-like restrictions such as player reach and fog of war. The rocket launch required Sol to manually ferry supplies between areas to compensate for failed late-game production connections. Developers tracking how agentic coding models handle long, complex tasks can follow these capability tests on daily.dev."}},{"@type":"Question","name":"How does GPT-5.6's game-playing ability compare to GPT-5.5 in Factorio?","acceptedAnswer":{"@type":"Answer","text":"GPT-5.6 Sol showed a clear improvement over GPT-5.5 in Factorio, successfully using transport belts and completing a full peaceful run through rocket launch. GPT-5.5, by contrast, failed to use transport belts at all during earlier attempts. Scaling issues still appeared for GPT-5.6, but later in the run than expected, during the post-blue-science production stage. Anyone comparing successive model releases for agentic reliability can keep up with these benchmarks on daily.dev."}}]} ```