Alibaba (Qwen)
Qwen2-72B benchmark results
8 results on 7 benchmarks from 5 sources.
Open weights Released: Jun 7, 2024 Updated Oct 22, 2026
Also listed as: qwen2-72b-instruct · Qwen2 Instruct (72B) · Qwen2-72B-32k · Qwen/Qwen2-72B-Instruct · Qwen/Qwen2-72B
All benchmark results
| Benchmarks | Score | Rank | Setting | Source type | Date |
|---|---|---|---|---|---|
| LMArena Text | 1262 | 313 | Overall, style control on; 37325 votes; 95% CI 1257–1267“qwen2-72b-instruct” | Official leaderboard | Oct 8, 2026 |
| Open LLM Leaderboard v2 | 43.6% | 58 | Average of 6 normalised benchmarks; bfloat16; 72.706B params“Qwen/Qwen2-72B-Instruct” | Official leaderboard | Mar 13, 2025 |
| GPQA Diamond | 40.8% | 236 | Epoch AI run, accuracy | Independent run | Jan 27, 2025 |
| MMLU | 82.4% | 11 | Exact match; 5-shot | Official leaderboard via Epoch AI | Jun 7, 2024 |
| MMLU-Pro | 55.6% | 153 | Overall accuracy; data source: TIGER-Lab; 72B“Qwen2-72B-32k” | Official leaderboard | Mar 11, 2026 |
| MATH Level 5 | 39.1% | 73 | Epoch AI run, accuracy | Independent run | Jan 27, 2025 |
| CMMLU | 89.7% | 4 | Five-shot average | Official leaderboard | Dec 6, 2024 |
| CMMLU | 87.5% | 5 | Zero-shot average | Official leaderboard | Dec 6, 2024 |
FAQ
Where do these Qwen2-72B scores come from?
From CMMLU leaderboard (GitHub README), Epoch AI Benchmarking Hub, LMArena leaderboard dataset, MMLU-Pro leaderboard (TIGER-Lab), and Hugging Face Open LLM Leaderboard v2 (archived). Each row links to the original leaderboard, dataset or report, and vendor-reported numbers are labelled.