← All benchmarks
Object/environment editing
We haven't written a description for this benchmark yet; see the card notes below for how each maker ran it.
- 1Nano Banana Pro · Google1102 Elo
How each maker reported it
- Nano Banana Pro p.4–5
Human side-by-side preference Elo (± CI omitted) as published in Google's model card, November 2025. Nano Banana Pro measured with Search on. Newer Google image models are not compared here.
Only each model's own maker's number is shown; scores a maker reports for rival models are left out.