bfl/flux-2-max Benchmark — 60/100 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.

MakerBlack Forest Labs

FamilyFLUX.2

Model sizenot disclosedundisclosed

Costfrom $0.07 / imagemid

Latency26.7s median

Run targetbfl/flux-2-max

Effective requestmodel: flux-2-max · size: 1024x1024 · seed: 42

SourcesBlack Forest Labs model docsBlack Forest Labs 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

62%

Typography Style33%

Writing accuracy83%

Spatial Reasoning

46%

Attributes Binding22%

Compositionality56%

Counting44%

Negation33%

Relative Position42%

Scale & Proportions22%

Human realism

51%

Faces & Expressions50%

Full Body33%

Hands58%

Multi-Subject67%

Professional Studio

61%

Camera & Lighting42%

Color Precision50%

Photorealism67%

Graphical design

77%

Data Visualisation67%

Layout & Design67%

Style Diversity67%

Truthfulness

79%

Photorealism67%

Physics & Reflections58%

World Knowledge75%