MGSM: leaderboard
A multilingual benchmark for mathematical questions.
Metric: Accuracy (%). Source: www.vals.ai. Status: saturated. 75 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.5 (20251101) (Thinking) | 95.2 |
| 2 | Claude Opus 4.5 (20251101) | 94.76 |
| 3 | Claude Opus 4.1 (20250805) (Thinking) | 94.44 |
| 4 | Claude Sonnet 4.5 (Thinking) | 94.33 |
| 5 | Claude Sonnet 4.5 20250929 (Thinking) | 94.33 |
| 6 | Claude Opus 4.1 (20250805) | 94.22 |
| 7 | GPT-5.2 (xHigh) | 94 |
| 8 | Gemini 3 Pro (Preview) (High) | 93.93 |
| 9 | Claude Opus 4 | 93.78 |
| 10 | Claude Opus 4 (20250514) | 93.78 |
| 11 | O4 Mini (2025-04-16) (High) | 93.42 |
| 12 | Gemini 3 Flash (Preview) (High) | 93.31 |
| 13 | Claude Sonnet 4 | 93.02 |
| 14 | Claude Sonnet 4 (20250514) | 93.02 |
| 15 | GPT-5.1 (High) | 92.98 |
Interactive version: theaggregate.ai/benchmark?slug=mgsm · How It Works · Data refreshed daily, snapshot 2026-09-05.