fal/bria/fast Benchmark — 48/100 Overall, 192 Prompts, Every Image Shown

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 run.

MakerBria

FamilyBria

Model sizenot disclosedundisclosed

Cost$0.028 / generationlow

Latency12.4s median

Run targetfal/bria/fast

Effective requestmodel: fast · size: 1024x1024 · seed: 42

SourcesBria model docsBria pricing

All 192 generations

Every challenge, grouped by category and subcategory. Solid-bordered tiles with a check passed the VLM judge; dashed tiles with a cross failed. Images are placeholders until generated.

Text Rendering

47%

Typography Style33%

Writing accuracy42%

Spatial Reasoning

58%

Attributes Binding56%

Compositionality56%

Counting44%

Negation56%

Relative Position33%

Scale & Proportions78%

Human realism

53%

Faces & Expressions42%

Full Body42%

Hands50%

Multi-Subject50%

Professional Studio

50%

Camera & Lighting42%

Color Precision33%

Photorealism33%

Graphical design

56%

Data Visualisation67%

Layout & Design67%

Style Diversity50%

Truthfulness

43%

Photorealism67%

Physics & Reflections42%

World Knowledge42%