fal/ideogram/v3 Benchmark — 49/100 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.

MakerIdeogram

FamilyIdeogram

Model sizenot disclosedundisclosed

Cost$0.06 / imagemid

Latency12.9s median

Run targetfal/ideogram/v3

Effective requestmodel: v3 · size: 1024x1024 · seed: 42

SourcesIdeogram model docsIdeogram 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

35%

Typography Style33%

Writing accuracy42%

Spatial Reasoning

52%

Attributes Binding44%

Compositionality56%

Counting56%

Negation44%

Relative Position25%

Scale & Proportions44%

Human realism

35%

Faces & Expressions17%

Full Body33%

Hands25%

Multi-Subject33%

Professional Studio

44%

Camera & Lighting33%

Color Precision42%

Photorealism33%

Graphical design

43%

Data Visualisation33%

Layout & Design33%

Style Diversity42%

Truthfulness

53%

Photorealism33%

Physics & Reflections42%

World Knowledge50%