local/bonsai-image-ternary-4b 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.

MakerBonsai

FamilyBonsai Image

Model size4B4B params

Costself-hostedfree

Latency4.1s median

Run targetlocal/bonsai-image-ternary-4b

Effective requestmodel: bonsai-image-ternary-4b · 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 accuracy17%

Spatial Reasoning

46%

Attributes Binding22%

Compositionality33%

Counting44%

Negation44%

Relative Position33%

Scale & Proportions44%

Human realism

49%

Faces & Expressions42%

Full Body50%

Hands50%

Multi-Subject50%

Professional Studio

30%

Camera & Lighting8%

Color Precision25%

Photorealism0%

Graphical design

46%

Data Visualisation33%

Layout & Design33%

Style Diversity25%

Truthfulness

44%

Photorealism100%

Physics & Reflections25%

World Knowledge42%