VMetaphor-Bench: leaderboard

Metric: MCQ accuracy (%; pooled over all 9,594 fidelity questions: two domain-presence questions per image, one structure-type, one meaning and the element-mapping questions; the generated image is judged by Qwen3.5-27B on multiple-choice questions (four options including cannot be determined) built from each metaphor's structured annotation; 1,500 visual metaphors curated from real creative imagery; conceptual prompts that state only the metaphorical idea). Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 11 models tracked.

Top models

#ModelScore
1GPT Image 1.585.4
2Nano Banana 2 (Gemini 3.1 Flash Image Preview)84.8
3Seedream 5.0 Lite78.5
4FLUX.2-dev76
5Qwen-Image-251274.9
6Janus-Pro-7B73
7Z-Image-Turbo72.4
8Stable Diffusion 3.5 Large70.8
9FLUX.1-dev68.4
10BAGEL-7B-MoT67

Interactive version: theaggregate.ai/benchmark?slug=vmetaphor-bench · How It Works · Data refreshed daily, snapshot 2026-09-26.