fal/krea/v2-medium Benchmark — 47/100 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.

MakerKrea

FamilyKrea

Model sizenot disclosedundisclosed

Cost$0.030 / imagelow

Latency18.5s median

Run targetfal/krea/v2-medium

Effective requestmodel: v2-medium · size: 1024x1024 · seed: 42

SourcesKrea model docsKrea 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

55%

Typography Style33%

Writing accuracy50%

Spatial Reasoning

58%

Attributes Binding67%

Compositionality33%

Counting33%

Negation67%

Relative Position42%

Scale & Proportions78%

Human realism

69%

Faces & Expressions50%

Full Body67%

Hands67%

Multi-Subject50%

Professional Studio

60%

Camera & Lighting50%

Color Precision58%

Photorealism67%

Graphical design

41%

Data Visualisation33%

Layout & Design56%

Style Diversity33%

Truthfulness

70%

Photorealism33%

Physics & Reflections58%

World Knowledge75%