CulMind - Calligraphy: leaderboard

Metric: Calligraphy subdomain score: macro-average of the task scores (0-100) in the subdomain, each of the 50 tasks keeping its own primary metric such as accuracy or F1, answer-only setting, images from more than 100 museum collections; higher is better. Source: arxiv.org. Saturation forecast: Around 2031. 14 models tracked.

Top models

#ModelScore
1Qwen 3 VL 8B Instruct65.1
2Gemini 3 Flash (Preview)62.3
3Qwen 3 VL 32B Instruct60.7
4Gemini 2.5 Flash56.4
5GPT-5.556.3
6InternVL3-14B53.7
7GLM-4.5V53
8InternVL3-8B51
9Qwen 3.5 9B50.5
10Qwen 3.5 27B50
11GPT-5.4 Mini47.1

Interactive version: theaggregate.ai/benchmark?slug=culmind-calligraphy · How It Works · Data refreshed daily, snapshot 2026-09-29.