AI for Education Pedagogy: leaderboard
One item is a four-option multiple-choice question from Chilean teacher qualification exams, translated into English, on topics like teaching strategies, assessment and classroom management. In this full cross-domain set, 899 of 920 questions are scored and the rest are fixed few-shot examples; answers are closed-book, single-turn and scored by exact letter match. 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 | GPT-6 Astra (High) | 92.66 | #6 (GPT-6 Astra) |
| 2 | GPT-6 Astra | 92.32 | #6 |
| 3 | Claude Fable 5.1 | 92.21 | #9 |
| 4 | GPT-5.5 | 92.1 | #31 |
| 5 | Gemini 3.8 Flash | 91.88 | #21 |
| 6 | Gemini 3.1 Pro (Preview) | 91.55 | #58 |
| 7 | Grok 4.6 | 91.55 | #26 |
| 8 | GPT-5.6 Sol | 91.21 | #19 |
| 9 | Claude Opus 5 | 91.1 | #12 |
| 10 | Gemini 3.6 Flash | 90.88 | #51 |
| 11 | Gemini 3.7 Flash | 90.77 | #23 |
| 12 | Gemini 3 Pro | 90.55 | #80 |
| 13 | Claude Fable 5 | 90.55 | #16 |
| 14 | Qwen 3.7 Max | 90.1 | #75 |
| 15 | Grok 4.5 | 89.66 | #48 |
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 · How It Works · Data refreshed daily, snapshot 2026-10-20.