AI for Education Pedagogy - Maths: leaderboard
The maths slice of the pedagogy exam set: four-option multiple-choice questions about how to teach and assess mathematics rather than how to solve it, taken from Chilean teacher qualification exams. Items are answered closed-book in a single turn and scored by exact match on the answer letter. Higher is better. Source: arxiv.org/abs/2506.18710.
Metric: Accuracy (%). Source: www.fab-ai.org. Saturation forecast: Estimated already saturated. 249 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Fable 5.1 | 96.03 | #9 |
| 2 | Claude Opus 5 | 95.24 | #12 |
| 3 | GPT-6 Astra | 95.24 | #6 |
| 4 | Gemini 3.6 Flash | 95.24 | #51 |
| 5 | Gemini 3.1 Pro (Preview) | 94.44 | #58 |
| 6 | Gemini 3.8 Flash | 94.44 | #21 |
| 7 | Claude Fable 5 | 94.44 | #16 |
| 8 | Muse Spark 1.2 | 94.44 | #62 |
| 9 | GPT-6 Astra (High) | 94.44 | #6 (GPT-6 Astra) |
| 10 | Gemini 3 Pro | 93.65 | #80 |
| 11 | Grok 4.6 | 93.65 | #26 |
| 12 | GPT-5.6 Sol | 92.86 | #19 |
| 13 | Gemini 3.5 Flash | 92.06 | #60 |
| 14 | Claude Sonnet 4.5 (Thinking) | 92.06 | #142 (Claude Sonnet 4.5) |
| 15 | Qwen 3.7 Max | 92.06 | #75 |
No result here: #3 Claude Opus 5.5, #7 GPT-6.1 Sol, #8 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=ai-for-education-pedagogy-maths · How It Works · Data refreshed daily, snapshot 2026-10-20.