ImageBench

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-1.1-pro

API

FLUX1.1 [pro] is Black Forest Labs' closed-weight flagship of the FLUX.1 generation, announced October 2024 alongside the BFL API, offering faster generation and improved prompt following over the original FLUX.1 [pro].

Maker
Black Forest Labs
Family
FLUX.1
Model Size
not disclosed
unknown
Cost
$0.04 / image
high
Run Target
bfl/flux-pro-1.1
Effective Request
width: 1024 · height: 1024 · output_format: png · seed: 7
47.0
Overall
46%
Capability
48.2
Est. Preference
88
Pass
104
Fail
4.6s
Avg Latency
3.0s
Min Latency
5.0s
Max Latency
Text Rendering53%Spatial Reasoning60%Human realism29%Truthfulness30%Professional Studio56%Graphical design46%Preference48%Latency55%

All 192 generations

Text Rendering53%

Typography Style100%

Writing accuracy42%

Spatial Reasoning60%

Attributes Binding89%

Compositionality67%

Counting56%

Negation67%

Relative Position58%

Scale & Proportions22%

Human realism29%

Faces & Expressions33%

Full Body17%

Hands50%

Multi-Subject0%

Truthfulness30%

Photorealism0%

Physics & Reflections50%

World Knowledge17%

Professional Studio56%

Camera & Lighting58%

Color Precision50%

Photorealism67%

Graphical design46%

Data Visualisation0%

Layout & Design33%

Style Diversity67%