EduResearchBench - Quantitative Research: leaderboard

Metric: Judge score (1-5; 5 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

#ModelScore
1GPT-53.12
2Gemini 3 Flash2.92
3Qwen 2.5 72B Instruct2.72
4Llama 3 70B Instruct2.51
5LongWriter-glm4-9B2.27

Interactive version: theaggregate.ai/benchmark?slug=eduresearchbench-quantitative-research · How It Works · Data refreshed daily, snapshot 2026-09-25.