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 ImageBench V1 run.

bfl/flux-1-dev

API

FLUX.1 [dev] is Black Forest Labs' 12B open-weight guidance-distilled model, released August 2024 under the FLUX.1 [dev] Non-Commercial License. Benchmarked here on the hosted BFL API endpoint rather than a local deployment.

Maker: Black Forest Labs
Family: FLUX.1
Model Size: 12B
high
Cost: $0.025 / image
high
Run Target: bfl/flux-dev
Effective Request: width: 1024 · height: 1024 · output_format: png · seed: 7
Sources: black-forest-labs/FLUX.1-dev model card BFL pricing

52.0

Overall

46%

Capability

58.1

Est. Preference

88

Pass

104

Fail

10.2s

Avg Latency

6.0s

Min Latency

34.0s

Max Latency

All 192 generations

Text Rendering53%

Typography Style67%

Writing accuracy50%

Spatial Reasoning54%

Attributes Binding78%

Compositionality56%

Counting44%

Negation44%

Relative Position42%

Scale & Proportions67%

Human realism26%

Faces & Expressions25%

Full Body0%

Hands58%

Multi-Subject17%

Truthfulness41%

Photorealism33%

Physics & Reflections58%

World Knowledge25%

Professional Studio70%

Camera & Lighting83%

Color Precision58%

Photorealism67%

Graphical design33%

Data Visualisation0%

Layout & Design11%

Style Diversity58%