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

KVBench (2026) tests whether text-to-image models can draw correct senior high-school biology 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 biology 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
1Nano Banana Pro (Gemini 3 Pro Image)77
2Gemini 2.5 Flash Image75.15
3GPT-Image (KVBench checkpoint unspecified)68.77
4Seedream 4.067.03
5FLUX 2 Max54.68
6FLUX.2-dev53.15
7Qwen-Image33.47
8HiDream (KVBench checkpoint unspecified)33
9SD3.5-Large (KVBench GGUF build, quantization unspecified)31.15
10FLUX.1-Krea-dev29.77
11Z-Image25.67
12Show-o218.59
13FLUX.1-dev15.25
14BAGEL-7B-MoT11.8#4668
15OmniGen28.27#4625

Interactive version: theaggregate.ai/benchmark?slug=kvbench-detailed-caption-biology-english · How It Works · Data refreshed daily, snapshot 2026-11-22.