MEGA-Bench Task - Ishihara Test — leaderboard

Metric: Task Score (%). Source: huggingface.co. 44 models tracked.

Top models

#ModelScore
1GPT-4o Mini58.6
2GPT-4o55.7
3Gemini 2.5 Pro44.3
4InternVL2.5-78B44.3
5InternVL3-78B42.9
6Ovis2-8B38.6
7Qwen 2 VL 7B37.1
8InternVL2-8B34.3
9InternVL3-14B31.4
10Gemini 1.5 Pro (002)31.4
11InternVL3-38B30
12Gemini 1.5 Flash (002)28.6
13Llama 4 Scout Base28.6
14Aquila-VL-2B25.7
15Gemma 3 27B (IT)22.9

Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-ishihara-test · How the rankings work · Data refreshed daily, snapshot 2026-07-22.