Mind's Eye - Mental Transformation: 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 recognizing rotated polycubes; higher is better.

Metric: Accuracy (%) on the mental transformation (polycube rotation) 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
1llava-v1.6-mistral-7B35.8#3999
2Gemini 2.5 Pro35.6#276
3GPT-4o32.7#751
4Idefics-8B (MindsEye checkpoint unspecified)32
5Phi-3.5-Vision Instruct (MindsEye checkpoint unspecified)29.3
6Llama 3.2 11B Vision29.2#4264
7MindsEye InternVL3-8B (checkpoint unspecified)29.1
8InstructBLIP-4B (MindsEye checkpoint unspecified)28.1
9Qwen 2.5 VL 32B Instruct27.6#1062
10GT-Thinker-7B26.7
11Qwen 2.5 VL 7B Instruct25.7#1902
12O325.1#237
13Vision-G1-7B25.1
14V-Thinker-8B (MindsEye checkpoint unspecified)24.5
15SmolVLM-2.2B (MindsEye checkpoint unspecified)23.5

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-mental-transformation · How It Works · Data refreshed daily, snapshot 2026-11-21.