Qwen2.5-Math-72B-Instruct: benchmark results
Alibaba's 72B open math specialist (September 2024) using chain-of-thought and tool-integrated reasoning, hitting 92.9 on MATH with TIR. Provider: Alibaba. Released 2024-09-19. Access: Open.
Aggregate Rating 1520 ± 20, rank #717 of 1656 rated models, from 67 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| KMP-Bench (Skills) - Follow-up Problem Generation | 86.3 | Pass rate (%) of generated follow-up problems and solutions | 100 |
| KMP-Bench (Skills) - Similar Problem Generation | 82.6 | Pass rate (%) of generated similar practice problems and sol | 100 |
| Open LLM Leaderboard - MATH Level 5 | 62.39 | Score | 99.9 |
| Open LLM Leaderboard - BBH | 64.52 | Score | 89.9 |
| Open LLM Leaderboard - MMLU-Pro | 48.12 | Score | 86.7 |
| Omni-MATH | 36.2 | Overall Accuracy (%) | 85.7 |
| MERA - SimpleAr | 99.8 | EM (%) | 84.8 |
| Open LLM Leaderboard - MuSR | 44.73 | Score | 81.9 |
| Open LLM Leaderboard - GPQA | 33.14 | Score | 81.1 |
| U-MATH - Sequences & Series | 68.83 | Accuracy (%) | 78.8 |
| U-MATH - Integral Calculus | 38.94 | Accuracy (%) | 77.3 |
| MERA - RWSD | 62.69 | Accuracy (%) | 76.4 |
Interactive version: theaggregate.ai/model?slug=qwen2-5-math-72b-instruct · How It Works · Data refreshed daily, snapshot 2026-10-20.