← Back to the main Geek Mode article
What was tested: the identical detailed prompt for a Hawaiian beach scene, given to Gemini 3.8 Flash (Flash Extended) and GPT-5.6 Sol (Extra High) in early September 2026. Outcome: both produced good images — one of the article’s balancing examples, not a case where either system clearly lost.
The prompt included a cat in the foreground, people, signs, food stands, a dog doing a specified action, and a number of environmental details. Both systems produced good images. ChatGPT’s version was slightly preferred, but that’s subjective — the Gemini image was attractive, coherent and absolutely usable.
That matters because the useful conclusion here is not “Gemini can’t make images.” It clearly can. A much worse Gemini image had turned up earlier in a complicated PDF workflow (a simple cartoon cat where something richer and photographic was called for), but that wasn’t a controlled comparison. This beach test was, and when the task was isolated and clear, both systems performed well.
← Previous: stopwatch · Back to the main Geek Mode article · Next: iterative UI editing →