OpenLearnLM - Pedagogical Knowledge: leaderboard
Pedagogical knowledge part of the OpenLearnLM Benchmark (Lee et al., 2026): four-option multiple-choice questions on teaching methods, learning theories and assessment from the Chilean teacher certification exam (CDPK) and Korea's KICE secondary-teacher exam, scored by exact-match accuracy in percent on the test split; higher is better.
Metric: Accuracy (%; four-option multiple choice from teacher certification exams). Source: arxiv.org. Saturation forecast: Estimated already saturated. 7 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro | 89.6 | #135 |
| 2 | GLM-4.7 | 87.5 | #365 |
| 3 | Kimi K2 (Thinking) | 87.1 | #435 (Kimi K2) |
| 4 | Claude Opus 4.5 | 86.1 | #152 |
| 5 | Grok 4.1 Fast | 86.1 | #426 |
| 6 | GPT-5.2 | 84.6 | #172 |
| 7 | DeepSeek V3.2 | 73.2 | #403 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=openlearnlm-pedagogical-knowledge · How It Works · Data refreshed daily, snapshot 2026-11-12.