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

KVBench (2026) tests whether text-to-image models can draw correct senior high-school physics 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 physics 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)76.2
2Gemini 2.5 Flash Image73.54
3GPT-Image (KVBench checkpoint unspecified)73.18
4Seedream 4.070.82
5FLUX.2-dev65.54
6FLUX 2 Max64.74
7SD3.5-Large (KVBench GGUF build, quantization unspecified)48.96
8Qwen-Image48.67
9FLUX.1-Krea-dev47.74
10HiDream (KVBench checkpoint unspecified)45.28
11Z-Image41.4
12FLUX.1-dev37.78
13BAGEL-7B-MoT29.38#4668
14OmniGen223.02#4625
15Show-o214.7

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