Mind's Eye - Visual Conceptual Slippage: 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 an odd-one-out task among six panels; higher is better.
Metric: Accuracy (%) on the visual conceptual slippage (odd one out among six panels) task of Mind's Eye (6-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
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Pro | 35.3 | #276 |
| 2 | PaliGemma-3B (MindsEye checkpoint unspecified) | 34.7 | |
| 3 | llava-v1.6-mistral-7B | 30.6 | #3999 |
| 4 | Qwen 2.5 VL 32B Instruct | 30.2 | #1062 |
| 5 | GT-Thinker-7B | 25.7 | |
| 6 | V-Thinker-8B (MindsEye checkpoint unspecified) | 25.5 | |
| 7 | GPT-4o | 25.2 | #751 |
| 8 | Vision-G1-7B | 24.1 | |
| 9 | O3 | 23.7 | #237 |
| 10 | MindsEye InternVL3-8B (checkpoint unspecified) | 23.7 | |
| 11 | Phi-3.5-Vision Instruct (MindsEye checkpoint unspecified) | 23.5 | |
| 12 | Llama 3.2 11B Vision | 23.1 | #4264 |
| 13 | Qwen 2.5 VL 7B Instruct | 22.7 | #1902 |
| 14 | SmolVLM-2.2B (MindsEye checkpoint unspecified) | 21.2 | |
| 15 | Qwen 2.5 VL 3B Instruct | 21 | #2713 |
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-visual-conceptual-slippage · How It Works · Data refreshed daily, snapshot 2026-11-21.