ImageBench V1 —

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.

Generation Details

Source-backed model context, size, cost, and request settings for this ImageBench V1 run.

fal/google/nano-banana-pro

API

Nano Banana Pro is Google DeepMind's higher-end image generation and editing model built on Gemini 3 Pro, designed for more accurate visuals, stronger reasoning, and better text rendering across Google products and APIs.

Maker: Google DeepMind
Family: Nano Banana Pro / Gemini image generation
Model Size: not disclosed
unknown
Cost: $0.15 / image
high
Run Target: fal/fal-ai/nano-banana-pro
Effective Request: aspect_ratio: 1:1 · resolution: 1K · num_images: 1 · output_format: png
Sources: Introducing Nano Banana Pro Google Cloud Nano Banana 2 and Nano Banana Pro GA Nano Banana Pro on fal

66.2

Overall

73%

Capability

59.1

Est. Preference

141

Pass

51

Fail

23.4s

Avg Latency

17.2s

Min Latency

82.3s

Max Latency

All 192 generations

Text Rendering100%

Typography Style100%

Writing accuracy100%

Spatial Reasoning89%

Attributes Binding100%

Compositionality89%

Counting78%

Negation89%

Relative Position100%

Scale & Proportions78%

Human realism33%

Faces & Expressions83%

Full Body8%

Hands8%

Multi-Subject33%

Truthfulness56%

Physics & Reflections42%

Photorealism33%

World Knowledge75%

Professional Studio89%

Camera & Lighting83%

Color Precision100%

Photorealism67%

Graphical design92%

Data Visualisation100%

Layout & Design78%

Style Diversity100%