AI for Education Visual Reasoning - match (process): leaderboard
Metric: Accuracy (%). Source: www.fab-ai.org. 67 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.6 Sol | 88.9 |
| 2 | Gemini 3.7 Flash | 88.9 |
| 3 | Gemini 3 Flash | 77.8 |
| 4 | Gemini 3.6 Flash | 77.8 |
| 5 | GPT-5.5 | 75 |
| 6 | Gemini 3.5 Flash | 75 |
| 7 | Kimi K3 | 75 |
| 8 | Gemini 3.1 Pro (Preview) | 72.2 |
| 9 | Qwen 3.6 Plus | 66.7 |
| 10 | GPT-5.4 (Medium) | 63.9 |
| 11 | GPT-5.4 | 61.1 |
| 12 | Qwen 3.5 397B A17B | 61.1 |
| 13 | Gemini 3 Pro | 58.3 |
| 14 | GPT-5.4 Mini (Medium) | 55.6 |
| 15 | Qwen 3.6 Flash | 55.6 |
Interactive version: theaggregate.ai/benchmark?slug=ai-for-education-visual-reasoning-match-process · How It Works · Data refreshed daily, snapshot 2026-09-05.