KVBench (Detailed Caption) - Geography (English): leaderboard
Metric: Checklist score (%) on the geography prompts in English, each given as a detailed caption that spells out the required visual content: KVBench knowledge-intensive text-to-image prompts from over 30 senior high-school textbooks (150 per subject and language, each with a textbook reference image); Qwen2.5-VL-32B-Instruct answers a checklist of four to six binary questions (key objects, attributes, spatial relations, reasoning outcomes) about each generated image, and the score is the mean share of satisfied items, in percent; higher is better. Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 16 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-Image (KVBench checkpoint unspecified) | 54.76 |
| 2 | Seedream-4.0 | 49.92 |
| 3 | FLUX.2-dev | 46.84 |
| 4 | Gemini 2.5 Flash Image | 38.95 |
| 5 | SD3.5-Large (KVBench GGUF build, quantization unspecified) | 34.32 |
| 6 | Qwen-Image | 33.38 |
| 7 | FLUX.2-max | 29.37 |
| 8 | FLUX.1-Krea-dev | 29.32 |
| 9 | HiDream (KVBench checkpoint unspecified) | 28.68 |
| 10 | Z-Image | 21.61 |
Interactive version: theaggregate.ai/benchmark?slug=kvbench-detailed-caption-geography-english · How It Works · Data refreshed daily, snapshot 2026-10-07.