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.

replicate/google/imagen-4

API

Imagen 4 is Google DeepMind's flagship text-to-image model, offered on Replicate as the standard-quality tier. It emphasizes fine detail rendering, strong prompt adherence, style versatility (photorealistic and abstract), and notably improved text/typography rendering, with output up to 2K resolution.

Maker: Google DeepMind
Family: Imagen 4
Model Size: not disclosed
unknown
Cost: $0.04 / image
high
Run Target: replicate/google/imagen-4
Effective Request: aspect_ratio: 1:1 · seed: 7
Sources: Imagen 4 — Replicate Announcing Imagen 4 family GA — Google Developers Blog