OpenAI launched ChatGPT Images 2.5 on September 8 with up to 50% lower latency than its predecessor and fixed the oversharpening and yellow color cast issues that affected previous models. Nano Banana 2, Google’s internal name for Gemini 3.1 Flash Image, and ChatGPT Images 2.5 faced off in six test categories: Nano Banana 2 won on lettering density, realism, and agentic research, while ChatGPT Images 2.5 won on spatial awareness, illustration, and abstract concepts. The verdict is essentially a tie, with differences stemming from specific minor errors such as a spelling mistake in a lettering-heavy scene or an incorrect date in a Bitcoin timeline infographic. Both models are judged roughly equivalent in overall aesthetic quality and performance.
Source: Read the original article
