AI for Education Pedagogy - Primary: leaderboard
The primary-school slice of the pedagogy exam set, holding the four-option multiple-choice questions tagged with that teaching level in the Chilean teacher qualification exams the benchmark is built from. 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 | GPT-5.5 | 96.71 | #31 |
| 2 | GPT-6 Astra (High) | 96.71 | #6 (GPT-6 Astra) |
| 3 | GPT-6 Astra | 96.24 | #6 |
| 4 | Gemini 3 Pro | 95.77 | #80 |
| 5 | Claude Fable 5.1 | 95.77 | #9 |
| 6 | Claude Opus 4.8 | 95.31 | #37 |
| 7 | Claude Opus 5 | 95.31 | #12 |
| 8 | Gemini 3.8 Flash | 95.31 | #21 |
| 9 | Claude Fable 5 | 94.84 | #16 |
| 10 | Gemini 3.6 Flash | 94.84 | #51 |
| 11 | Gemini 3.7 Flash | 94.84 | #23 |
| 12 | Grok 4.6 | 94.84 | #26 |
| 13 | GPT-5.6 Sol | 94.37 | #19 |
| 14 | Qwen 3.6 Max Preview | 94.37 | #85 |
| 15 | Gemini 2.5 Pro | 93.9 | #152 |
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-primary · How It Works · Data refreshed daily, snapshot 2026-10-20.