Topology Test: Grok Imagine 2.0 Beats ChatGPT by Correctly Understanding Genus

luismbat · x · 2026-08-10

A developer tested Grok Imagine 2.0 against other image models using the prompt: "A mathematician drinking from an object of genus 1."

The results showed that ChatGPT fell into the common analogy trap of "donut equals genus 1," turning it into a mug and accidentally creating an object of genus 2. In contrast, Grok provided the correct answer: a regular coffee mug already has a genus of 1.

Original post →

More from Fun

Fun channel →