Qwen 2 72B: benchmark results
Alibaba's open 72B flagship of the 2024 Qwen2 series, with a 128K context via YaRN. Provider: Alibaba. Released 2024-06-06. Access: Open.
Aggregate Rating 1578 ± 12, rank #521 of 1656 rated models, from 232 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenEvals - GSM8K | 89.5 | Accuracy (%) | 100 |
| RABBITS B4B | 83.72 | Accuracy (%) | 100 |
| Open LLM Leaderboard v1 - GSM8K | 85.29 | Accuracy (%) (5-shot) | 99.9 |
| Open LLM Leaderboard v1 - MMLU | 83.79 | Accuracy (%) (5-shot) | 99.8 |
| Open Chinese LLM - C-Eval Semantic | 93.47 | Accuracy (%) | 99.7 |
| Open Chinese LLM - CMMLU | 78.66 | Accuracy (%) | 99.7 |
| Open Chinese LLM - GSM8K | 75.36 | Accuracy (%) | 99.7 |
| Open PL LLM - PolEmo2-OUT (generative, 5-shot) | 82.59 | Accuracy (%) | 99.4 |
| Open LLM Leaderboard - MMLU-Pro | 57.31 | Score | 99.2 |
| Open Arabic LLM - Arabic MMLU HT Global Facts | 54 | Accuracy (%) | 98.8 |
| Open PL LLM - PolEmo2-OUT (multiple choice, 5-shot) | 82.19 | Accuracy (%) | 98.7 |
| Open Chinese LLM Leaderboard | 72.66 | Official Nonzero-Task Mean (%) | 98.5 |
Interactive version: theaggregate.ai/model?slug=qwen-2-72b · How It Works · Data refreshed daily, snapshot 2026-10-20.