GeoReader MindCube - Overall: leaderboard
GeoReader MindCube - Overall. MindCube spatial reasoning;source table keeps backbone groups,geometry reader/feature alignment and full-method checkpoint distinct. Unspecified DeepSeek-Flash endpoint remains unspecified. Metric: Accuracy (%). Primary published table; source configurations remain distinct.
Metric: Accuracy (%). Source: iris0329.github.io. Saturation forecast: Around October 2027. 28 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | O3 | 56.6 | #123 |
| 2 | MiMo-V2.6-Flash | 54.5 | #62 |
| 3 | Gemini 2.5 Pro | 52.2 | #150 |
| 4 | Seed-1.6 | 46.1 | #259 |
| 5 | GPT-4.1 | 45.1 | #243 |
| 6 | Qwen 2.5 VL 7B | 34.8 | #632 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #7 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=georeader-mindcube-overall · How It Works · Data refreshed daily, snapshot 2026-10-16.