Qwen2.5-Math-7B: benchmark results
Alibaba's Qwen2.5-Math-7B base model, specialized for solving English and Chinese math problems via chain-of-thought and tool-integrated reasoning, upgraded from the Qwen2-Math series. Provider: Alibaba. Released 2024-09-19. Access: Open.
Aggregate Rating 1301 ± 17, rank #1482 of 1656 rated models, from 66 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard - MATH Level 5 | 30.51 | Score | 82.2 |
| Open Korean LLM - KorNAT-Helpful | 49.23 | Normalized accuracy (%) | 77.8 |
| EuroEval Spanish NLU - Sentiment Headlines ES | 40.68 | Matthews correlation coefficient (%) | 54.1 |
| Open LLM Leaderboard - GPQA | 29.36 | Score | 48.7 |
| EuroEval Spanish NLU - ScaLA ES | 9.62 | Matthews correlation coefficient (%) | 45.3 |
| EuroEval Spanish NLU | 39.78 | NLU Average Score (%) | 42.4 |
| EuroEval Portuguese NLU - ScaLA PT | 7.68 | Matthews correlation coefficient (%) | 40.7 |
| EuroEval Norwegian Knowledge - Idioms NO | 1.38 | MCC (x100) | 40.4 |
| EuroEval Spanish | 34.09 | Average Score (%) | 39.7 |
| EuroEval Portuguese | 37.62 | Average Score (%) | 39.2 |
| EuroEval Portuguese NLU - SST-2 PT | 72.07 | Matthews correlation coefficient (%) | 38.2 |
| EuroEval Norwegian Common Sense Reasoning | 29.52 | Common Sense Reasoning Average Score (%) | 38.1 |
Interactive version: theaggregate.ai/model?slug=qwen2-5-math-7b · How It Works · Data refreshed daily, snapshot 2026-10-20.