Mind's Eye - Symmetric Structures: leaderboard

Mind's Eye (2026) is a multiple-choice benchmark of visuospatial reasoning tasks adapted from human intelligence tests, with procedurally generated items answered with chain-of-thought prompting. This board is accuracy in percent on reasoning about symmetric structures; higher is better.

Metric: Accuracy (%) on the symmetric structures task of Mind's Eye (4-option multiple choice, 100 procedurally generated items, chain-of-thought prompt, answers extracted by an LLM evaluator; keys uniformly distributed over the options); higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 18 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 2.5 Pro31.4#276
2Phi-3.5-Vision Instruct (MindsEye checkpoint unspecified)29.7
3Llama 3.2 11B Vision29.3#4264
4MindsEye InternVL3-8B (checkpoint unspecified)29.1
5Qwen 2.5 VL 32B Instruct26.3#1062
6Idefics-8B (MindsEye checkpoint unspecified)25
7llava-v1.6-mistral-7B24.5#3999
8GT-Thinker-7B24.1
9Vision-G1-7B23.6
10V-Thinker-8B (MindsEye checkpoint unspecified)22.9
11O322.3#237
12Qwen 2.5 VL 3B Instruct21.2#2713
13Qwen 2.5 VL 7B Instruct20.2#1902
14GPT-4o19.1#751
15BLIP-2-OPT-2.7B17#6196

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=mind-s-eye-symmetric-structures · How It Works · Data refreshed daily, snapshot 2026-11-21.