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/bria/fast

API

BRIA Fast is BRIA AI's latency-oriented text-to-image model, promoted for commercially safe generation from models trained exclusively on licensed data and served here through fal.

Maker: BRIA AI
Family: BRIA Text-to-Image
Model Size: not disclosed
unknown
Cost: $0.028 / generation
high
Run Target: fal/fal-ai/bria/text-to-image/fast
Effective Request: aspect_ratio: 1:1 · num_images: 1 · seed: 7
Sources: BRIA official site BRIA Fast on fal

47.9

Overall

51%

Capability

44.8

Est. Preference

98

Pass

94

Fail

12.4s

Avg Latency

7.0s

Min Latency

23.6s

Max Latency

All 192 generations

Text Rendering33%

Typography Style67%

Writing accuracy25%

Spatial Reasoning68%

Attributes Binding100%

Compositionality89%

Counting56%

Negation78%

Relative Position58%

Scale & Proportions33%

Human realism19%

Faces & Expressions50%

Full Body0%

Hands0%

Multi-Subject33%

Truthfulness37%

Photorealism33%

Physics & Reflections58%

World Knowledge17%

Professional Studio81%

Camera & Lighting75%

Color Precision83%

Photorealism100%

Graphical design58%

Data Visualisation0%

Layout & Design33%

Style Diversity92%