Someone used Claude Code to make a sailing sim, then challenged other frontier LLMs to a race

Published Nov 7, 2026, 2:00 AM EST Simon is a Computer Science BSc graduate who has been writing about technology since 2014, and using Windows machines since 3.1. After working for an indie game studio and acting as the family's go-to technician for all computer issues, he found his passion for writing and decided to use his skill set to write about all things tech. Since beginning his writing career, he has written for many different publications such as WorldStart, Listverse, and MakeTechEasier. However, after finding his home at MakeUseOf in February 2019, he would eventually move on to its sister site, XDA, to bring the latest and greatest in Windows, Linux, and DIY electronics. Sign in to your XDA account Summary Claude Code sim lets LLMs helm virtual yachts using world-state inputs every 5s. Gemini 4 Argon won easily; GPT-6 Astra 2nd, Claude Opus stalled, Qwen visited a beach. The leading theory says Gemini's spatial awareness (for Maps/Earth) gave it an edge over the others. If you gave Claude Opus 5.5, GPT-6 Astra, and Gemini 4 Argon a yacht each, then told them to sail across a body of water and back with weather conditions, who do you think will win? While it'd be a very expensive project to do (both in terms of getting the boats and feeding each LLM with credits), someone has done the next best thing. They used Claude Code to make a realistic boat-racing simulator, then tasked the frontier LLM models with a virtual race. Claude Code created a sailing arena for four AI models to duke it out The results were wildly varied Over on the ClaudeAI subreddit, user abhi3188 showed off what they made in Claude Code. It's a sailing simulator that allows an LLM to take the helm of a virtual yacht and sail back and forth across a body of water. As for the sailors, abhi3188 picked Claude Opus 5.5, GPT-6 Astra, Gemini 4 Argon, and Qwen 3.5-9B, the latter of which they admit they "threw it in there for fun." Here's how the boat race worked: Every 5 secs each model gets the full state of the world (23 raw numbers like heading, speed, wind, where the next mark is etc) and sets its helm and sails. No sailing tips. The sim pauses while it thinks, so slow models aren't punished. As for the race, Qwen came in dead last as it seemingly refused to race and instead visited a nearby beach. Despite having a hand in coding the sim, Claude Opus 5.5 came in third after it stalled on the way home and timed out. GPT-6 Astra finished second after completing a full lap, and Gemini 4 Argon secured an easy win, at 1/5th of the cost in tokens versus Astra. People speculated as to why Gemini was the superior sailor here, and a leading theory is that Google gave it superior spatial awareness because they also want to use it in Google Maps and Earth. Either way, I think I'd enjoy more of these AI duels. Battlebots, anyone?