Multimodal benchmark

# MindCube leaderboard

> MindCube results for 2 AI models, led by Gemma 3 12B at 46.7%. What the benchmark measures, who runs it, and a source for every score.
- Canonical page: https://noometry.com/benchmarks/mindcube
- Last updated: 2026-10-10
- Title: MindCube Leaderboard (October 2026): Scores by Model

As of October 2026, Gemma 3 12B has the highest published MindCube score on Noometry at 46.7%, out of 2 models with results.

Last verified October 10, 2026

## About MindCube

Spatial mental modeling from a small number of images.

- **Category:** [Multimodal](https://noometry.com/best/multimodal)
- **Introduced:** 2025
- **Unit:** Percent (random guessing ≈ 0%)
- **Official site:** [mind-cube.github.io](https://mind-cube.github.io)

## Top 2 models

Top models on MindCube

1.  Gemma 3 12B 46.7%
2.  Claude Sonnet 4 44.8%
3.  44.044.545.045.546.046.547.0

Sponsored placements are available on pages like this one. [Advertise on Noometry](https://noometry.com/advertise)

## All results

MindCube results by model
| # | Model | Provider | Score | Setting | Source | Date |
| --- | --- | --- | --- | --- | --- | --- |
| 1 | [Gemma 3 12B](https://noometry.com/models/gemma-3-12b) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 46.7% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 2 | [Claude Sonnet 4](https://noometry.com/models/claude-sonnet-4) | [Anthropic](https://noometry.com/providers/anthropic) | 44.8% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |

## Compare the leaders

-   [Gemma 3 12B vs Claude Sonnet 4](https://noometry.com/compare/claude-sonnet-4-vs-gemma-3-12b)

## Other multimodal benchmarks

-   [LMArena Vision](https://noometry.com/benchmarks/arena-vision)
-   [Video-MME](https://noometry.com/benchmarks/video-mme)
-   [GeoBench](https://noometry.com/benchmarks/geobench)
-   [VPCT](https://noometry.com/benchmarks/vpct)
-   [Blueprint-Bench 2](https://noometry.com/benchmarks/blueprint-bench-2)
-   [Furniture Assembly](https://noometry.com/benchmarks/furniture-assembly)
-   [LMArena Document](https://noometry.com/benchmarks/arena-document) (reference)
-   [ScienceQA](https://noometry.com/benchmarks/scienceqa) (reference)
-   [SpatialViz-Bench](https://noometry.com/benchmarks/spatialviz-bench) (reference)

## Frequently asked questions

### What does MindCube measure?

Spatial mental modeling from a small number of images.

### Which model has the highest MindCube score?

As of October 2026, Gemma 3 12B has the highest published MindCube score on Noometry at 46.7%, out of 2 models with results.

### What is the best open-weight model on MindCube?

Gemma 3 12B has the highest MindCube accuracy among open-weight models at 46.7%, ranking 1 of 2 overall.

### Cite this page

Noometry. (2026). MindCube leaderboard. Retrieved October 10, 2026, from https://noometry.com/benchmarks/mindcube

Quote Noometry with a link back to this page. It is also available in [Markdown](https://noometry.com/md/benchmarks/mindcube.md).
