Project AURIONAI Use-Cases Research & Innovation
Alibaba (Qwen)

Qwen-Image-Bench

Qwen vendor image benchmark; methodology unreleased, chart only

generative mediastatic
5scores
5models
0independent
0maintainer-run
5vendor-reported

Definition

Version
not in the notes
Category
generative media
Maintainer
Alibaba (Qwen)
Status
static
Home page
not in the notes

Sources

  1. Qwen-Image-2.1 release: Qwen's 7B open-weights image model datanorth.ai https://datanorth.ai/news/qwen-releases-qwen-image-2-1

Scores at a Glance

Where this comes from
Each bar is one model, with its most independent recorded score. Hatched bars are vendor-reported; the word beside each bar says who ran it. Settings and dates differ between rows; the Scores table below has every result.

Scores

5 of 5 scores shown qwen-image-bench-scores.csv
ModelScoreWho ran itSettingView or populationDateSource
OpenAIGPT Image 2.5 Sunburst67.01 scorevendor-reportedQwen chart only; methodology unreleasedQwen vendor chartSep 2026datanorth.ai
Alibaba (Qwen)Qwen-Image-3.0-Pro62.36 scorevendor-reportedQwen chart only; methodology unreleasedQwen vendor chartSep 2026datanorth.ai
Alibaba (Qwen)Qwen-Image-2.160.28 scorevendor-reportedQwen chart only; methodology unreleasedQwen vendor chartSep 2026datanorth.ai
GoogleNano Banana 2 (Gemini 3.1 Flash Image)59.82 scorevendor-reportedQwen chart only; methodology unreleasedQwen vendor chartSep 2026datanorth.ai
Black Forest LabsFLUX 2 Max55.33 scorevendor-reportedQwen chart only; methodology unreleasedQwen vendor chartSep 2026datanorth.ai
Where this comes from
Scores from different settings, harnesses or dates are not directly comparable; the row says which. Sorted by score, highest first, as a reading aid, not a ranking.

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "Qwen-Image-Bench", https://aurion.semai.id/benchmarks/qwen-image-bench/, records of October 2026.