MindCube leaderboard
As of 6 Nov 2026, the top score on MindCube is 46.7% by Gemma 3 12B (Google), ahead of mPLUG-Owl3-7B-241101 at 44.9%, across 5 model configurations with a published result (Epoch AI Benchmarking Hub, result dated 12 Mar 2025). Questions that require building a mental map from several views of a scene. MindCube is shown for reference only and does not count toward the BenchLeader Index.
Key facts (as of 6 Nov 2026)
- Gemma 3 12B holds the top MindCube score at 46.7% (Epoch AI Benchmarking Hub, 12 Mar 2025).
- mPLUG-Owl3-7B-241101 is second on MindCube at 44.9% (Epoch AI Benchmarking Hub, 26 Nov 2024).
- MindCube: Spatial reasoning from multiple views (Epoch AI Benchmarking Hub).
- 5 model configurations have a published MindCube result on BenchLeader (BenchLeader, 6 Nov 2026).
- MindCube does not feed the BenchLeader Index; it is shown for reference (BenchLeader methodology).
Spatial reasoning from multiple views.
- Published by
- Epoch AI Benchmarking Hub
- Category
- Multimodal
- Index weight
- Reference only
- Models
- 5
- Data as of
- 6 Nov 2026
CC BY 4.0 — Epoch AI, ‘AI Benchmarking Hub’, epoch.ai/benchmarks. Mirrored boards credit their original publishers.
What the test looks like
Questions that require building a mental map from several views of a scene.
How it is scored
Overall score, as published.
What to keep in mind
Very few models measured.
- 1Gemma 3 12B46.7%
- 2mPLUG-Owl3-7B-24110144.9%
- 3Claude Sonnet 444.8%
- 4LLaVA-Video-7B-Qwen242.0%
- 5LongVA-7B29.5%
| # | ||||
|---|---|---|---|---|
| 1 |  Gemma 3 12Bopen ↗ | 46.7% | 37.9 | 2025-03-12 |
| 2 |  mPLUG-Owl3-7B-241101 | 44.9% | – | 2024-11-26 |
| 3 |  Claude Sonnet 4 | 44.8% | 50.5 | 2025-05-22 |
| 4 |  LLaVA-Video-7B-Qwen2 | 42.0% | – | 2024-09-02 |
| 5 | LongVA-7B | 29.5% | – | 2024-06-13 |
More multimodal benchmarks · All LLM benchmarks
Related benchmarks
Cite as: BenchLeader, “MindCube leaderboard”, data as of 6 Nov 2026, https://www.benchleader.com/benchmarks/mindcube.
MindCube: questions
- What does MindCube measure?
- Questions that require building a mental map from several views of a scene. Scores are reported in percent of tasks solved; higher is better.
- What is the top score on MindCube?
- As of 6 Nov 2026, the highest score on MindCube is 46.7% by Gemma 3 12B (result dated 12 Mar 2025), ahead of mPLUG-Owl3-7B-241101 at 44.9%. Scores come from Epoch AI Benchmarking Hub and are re-read daily.
- Which AI model leads MindCube?
- Gemma 3 12B leads MindCube with 46.7% as of 6 Nov 2026 (result dated 12 Mar 2025), ahead of mPLUG-Owl3-7B-241101 at 44.9%.
- How many models have MindCube results?
- 5 model configurations have a MindCube result on BenchLeader, all taken from Epoch AI Benchmarking Hub.
- Who runs MindCube and how often is it updated?
- MindCube is published by Epoch AI Benchmarking Hub. BenchLeader re-reads the published results once a day and records the date each result was published.
- Does MindCube count toward the BenchLeader Index?
- No. MindCube is shown for reference but left out of the composite index.