Bongard Problems (Classic): leaderboard
Metric: Accuracy (% correct). Source: github.com. 15 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 55.4 |
| 2 | Gemini 3 Pro (Preview) | 50.5 |
| 3 | Gemini 3 Flash (Preview) | 38.2 |
| 4 | Gemini 2.5 Flash | 16.2 |
| 5 | Mistral Medium 3.1 | 11.5 |
| 6 | Ministral 3 14B | 9.8 |
| 7 | Mistral Medium 3 | 9.3 |
| 8 | Ministral 3 8B | 8.6 |
| 9 | Mistral Small 3.2 | 8.3 |
| 10 | Mistral Medium 3.5 | 6.9 |
| 11 | Magistral Small 1.2 | 6.6 |
| 12 | Magistral Medium 1.2 | 6.6 |
| 13 | Mistral Small 4 | 5.6 |
| 14 | Mistral Large 3 | 5.1 |
Interactive version: theaggregate.ai/benchmark?slug=bongard-problems-classic · How It Works · Data refreshed daily, snapshot 2026-09-08.