EduResearchBench - Qualitative Research: leaderboard
EduResearchBench (Yue et al., 2026) qualitative research module: four atomic tasks that write thick description, the analysis process, findings and a trustworthiness statement for an education study, in Chinese and English. Doubao-seed-1.6 and DeepSeek-V3.2 each write a critique and a 1-5 score, and the two scores are averaged; higher is better.
Metric: Judge score (1-5; 4 atomic tasks, Chinese and English, mean of Doubao-seed-1.6 and DeepSeek-V3.2 critique-then-score ratings). Source: arxiv.org. Saturation forecast: Around February 2027. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 | 3.65 | #163 |
| 2 | Gemini 3 Flash | 3.45 | #159 |
| 3 | Qwen 2.5 72B Instruct | 3.01 | #1093 |
| 4 | Llama 3 70B Instruct | 2.67 | #1684 |
| 5 | LongWriter-glm4-9B | 2.55 | #2419 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=eduresearchbench-qualitative-research · How It Works · Data refreshed daily, snapshot 2026-11-12.