192 evaluations across 6 categories
Benchmark V1 verdicts are produced by VLM judges and can contain mistakes. Treat PASS/FAIL labels as machine-assisted assessments, and inspect the images yourself. Learn more about the methodology.
Source-backed model context, size, cost, and request settings for this ImageBench V1 run.
Nano Banana 2 is Google's fast Gemini image generation and editing model, exposed here through fal, with emphasis on high-quality image creation, grounded world knowledge, and practical generation speed.
GPT Image 2.5 Sunburst is one of two API variants OpenAI released alongside ChatGPT Images 2.5 (September 2026), built for workflows where editing precision matters most.
On ImageBench V1 (192 prompts, 6 categories), Google Nano Banana 2 scores 74.9 overall with a 77.1% task pass rate, while GPT Image 2.5 Sunburst scores 82.1 overall with a 88% pass rate. The biggest difference is text rendering: Google Nano Banana 2 passes 73.3% of those prompts vs 100% for GPT Image 2.5 Sunburst. Google Nano Banana 2 is an API model ($0.08 / image at 1K); GPT Image 2.5 Sunburst is an API model (token-based).
GPT Image 2.5 Sunburst is better overall on ImageBench V1, scoring 82.1 vs 74.9. Google Nano Banana 2 is strongest at graphic design (100%). GPT Image 2.5 Sunburst is strongest at text rendering (100%).
Yes — on the Text Rendering category of ImageBench V1, Google Nano Banana 2 passes 73.3% of prompts and GPT Image 2.5 Sunburst passes 100%. That is the largest per-category difference between Google Nano Banana 2 and GPT Image 2.5 Sunburst.
The ImageBench gallery shows Google Nano Banana 2 and GPT Image 2.5 Sunburst outputs for the same 192 prompts side by side, filterable by category and difficulty — the fastest way to judge them with your own eyes instead of trusting aggregate scores. Open them in the gallery.