3,320 arithmetic word problems from various sources, testing basic arithmetic reasoning.
Accuracy is the reported evaluation metric for MAWPS. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.
Higher is better
| Rank | Model | Trust | Score | Year | Links | Edit |
|---|---|---|---|---|---|---|
| 01 | gpt-4o | paper | 97.2 | 2025 | Source ↗ | Edit result |
| 02 | claude-35-sonnet | paper | 95.8 | 2025 | Source ↗ | Edit result |
| 03 | Claude 3.5 Sonnet | paper | 95.8 | 2025 | Source ↗ | Edit result |
| 04 | llama-3-70b | paper | 94.1 | 2025 | Source ↗ | Edit result |
| 05 | Llama 3 70B | unverified | 94.1 | 2025 | Source ↗ | Edit result |