local/z-image-turbo-6b 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.

MakerTongyi

FamilyZ-Image

Model size6B6B params

Costself-hostedfree

Latency18.1s median

Run targetlocal/z-image-turbo-6b

Effective requestmodel: z-image-turbo-6b · size: 1024x1024 · seed: 42

SourcesModel cardWeights & checkpoint

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

43%

Typography Style33%

Writing accuracy50%

Spatial Reasoning

56%

Attributes Binding44%

Compositionality56%

Counting44%

Negation56%

Relative Position58%

Scale & Proportions78%

Human realism

55%

Faces & Expressions42%

Full Body42%

Hands42%

Multi-Subject50%

Professional Studio

57%

Camera & Lighting42%

Color Precision33%

Photorealism67%

Graphical design

34%

Data Visualisation33%

Layout & Design11%

Style Diversity25%

Truthfulness

46%

Photorealism0%

Physics & Reflections33%

World Knowledge33%