KVBench (Detailed Caption) - Geography (English): leaderboard

KVBench (2026) tests whether text-to-image models can draw correct senior high-school geography content, such as diagrams and scientific illustrations, from English prompts given as a detailed caption that spells out the required visual content. Qwen2.5-VL-32B checks each image against a list of textbook-grounded constraints; the score is the percentage satisfied, higher is better.

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. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1GPT-Image (KVBench checkpoint unspecified)54.76
2Seedream 4.049.92
3FLUX.2-dev46.84
4Gemini 2.5 Flash Image38.95
5SD3.5-Large (KVBench GGUF build, quantization unspecified)34.32
6Qwen-Image33.38
7FLUX 2 Max29.37
8FLUX.1-Krea-dev29.32
9HiDream (KVBench checkpoint unspecified)28.68
10Z-Image21.61
11Show-o215.21
12Nano Banana Pro (Gemini 3 Pro Image)15.11
13FLUX.1-dev12.35
14BAGEL-7B-MoT12.04#4619
15OmniGen211.84#4579

Interactive version: theaggregate.ai/benchmark?slug=kvbench-detailed-caption-geography-english · How It Works · Data refreshed daily, snapshot 2026-11-21.