MCLM - MT-AIME2024: leaderboard
Each model solves the 30 AIME 2024 problems machine-translated by GPT-4o into 55 languages (1,650 problems per model), with temperature 0 and at most 3,276 output tokens, checked by a rule-based answer verifier; the score is the mean of the 55 per-language accuracies. It is the competition-level sibling of MCLM's MT-MATH100 subset and differs from English AIME 2024 boards by averaging over languages; the short output cap limits reasoning models.
Metric: Accuracy (%). Source: arxiv.org. Saturation forecast: Estimated already saturated. 5 models tracked. Updated 17 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | O3 Mini (2025-01-31) | 45.33 | #498 |
| 2 | GPT-4o Mini (2024-07-18) | 20.18 | #1494 |
| 3 | Qwen2.5-Math-7B-Instruct | 20.12 | #8400 |
| 4 | DeepSeek R1 Distill Qwen 1.5B | 17.21 | #9148 |
| 5 | Qwen2.5-Math-1.5B-Instruct | 16.36 | #9038 |
No result here: #1 Claude Fable 5.2, #2 GPT-6.2 Astra, #3 Gemini 4 Carbon.
Interactive version: theaggregate.ai/benchmark?slug=mclm-mt-aime2024 · How It Works · Data refreshed daily, snapshot 2026-11-17.