MGSM (Aggregate)
Multilingual Grade School Math: aggregate accuracy across 10 languages (Spanish, French, German, Russian, Chinese, Japanese, Thai, Swahili, Bengali, Telugu). Score = % of math word problems solved correctly. These entries are OpenAI-published model results, not an independent cross-provider leaderboard.
| # | Model | Score | Evidence | Tested |
|---|---|---|---|---|
| 1 | OpenAI: o4 Mini | 93.7 | model-only official_provider | 2026-06-07 |
| 2 | Meta: Llama 4 Maverick | 92.3 | model-only independent_benchmark | 2000-01-01 |
| 3 | OpenAI: o3 | 92.3 | model-only official_provider | 2026-06-07 |
| 4 | Meta: Llama 3.3 70B Instruct | 91.1 | model-only independent_benchmark | 2000-01-01 |
| 5 | OpenAI: o3 Mini | 90.8 | model-only official_provider | 2026-06-07 |
| 6 | Meta: Llama 4 Scout | 90.6 | model-only independent_benchmark | 2000-01-01 |
| 7 | OpenAI: o1 | 89.3 | model-only official_provider | 2026-06-07 |
| 8 | OpenAI: GPT-4.1 | 86.9 | model-only official_provider | 2026-06-07 |
| 9 | Qwen: Qwen3 235B A22B | 83.5 | model-only independent_benchmark | 2000-01-01 |
Multilingual Grade School Math: aggregate accuracy across 10 languages (Spanish, French, German, Russian, Chinese, Japanese, Thai, Swahili, Bengali, Telugu). Score = % of math word problems solved correctly. These entries are OpenAI-published model results, not an independent cross-provider leaderboard.
This benchmark contributes direct public evidence. Read its scope before generalizing the result.
A win here is a win on MGSM (Aggregate). Broad task pages require independent corroboration before naming a general winner.