New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
MT-AIME 2024 benchmark
AI model leaderboard for the MT-AIME 2024 benchmark. Compare how large language models score on MT-AIME 2024, see the full ranking, and understand what this AI benchmark measures. Qwen3-235B-A22B currently leads with 80.8. Multilingual AIME 2024 — translations of AIME 2024 problems into multiple languages, tests cross-lingual math.
Leaderboard
| # | Model | Organization | Score | Variant | Source |
|---|---|---|---|---|---|
| #1 | Qwen3-235B-A22B | Qwen | 80.8 | thinking-mt-2024 | official ↗ |
| #2 | Qwen3-32B | Qwen | 75 | thinking-mt-2024 | official ↗ |
| #3 | Qwen3-30B-A3B | Qwen | 73.9 | thinking-mt-2024 | official ↗ |
| #4 | Qwen3-14B | Qwen | 73.3 | thinking-mt-2024 | official ↗ |
| #5 | Qwen3-8B | Qwen | 65.4 | thinking-mt-2024 | official ↗ |
| #6 | Qwen3-4B | Qwen | 60.7 | thinking-mt-2024 | official ↗ |
| #7 | Qwen3-1.7B | Qwen | 36.1 | thinking-mt-2024 | official ↗ |
| #8 | Qwen3-235B-A22B | Qwen | 32.4 | non-thinking-mt-2024 | official ↗ |
| #9 | Qwen3-30B-A3B | Qwen | 24.6 | non-thinking-mt-2024 | official ↗ |
| #10 | Qwen3-32B | Qwen | 24.1 | non-thinking-mt-2024 | official ↗ |
| #11 | Qwen3-14B | Qwen | 23.2 | non-thinking-mt-2024 | official ↗ |
| #12 | Qwen3-8B | Qwen | 16.6 | non-thinking-mt-2024 | official ↗ |
| #13 | Qwen3-4B | Qwen | 13.9 | non-thinking-mt-2024 | official ↗ |
| #14 | Qwen3-0.6B | Qwen | 7.8 | thinking-mt-2024 | official ↗ |
| #15 | Qwen3-1.7B | Qwen | 4.9 | non-thinking-mt-2024 | official ↗ |
| #16 | Qwen3-0.6B | Qwen | 1.5 | non-thinking-mt-2024 | official ↗ |
Frequently asked questions about MT-AIME 2024
What is the MT-AIME 2024 benchmark?
Multilingual AIME 2024 — translations of AIME 2024 problems into multiple languages, tests cross-lingual math.
How is the MT-AIME 2024 benchmark scored?
MT-AIME 2024 is scored using the accuracy metric, where a higher score is better. GenAIList aggregates reported scores from model providers and papers into a single ranked leaderboard.
Which AI model scores highest on MT-AIME 2024?
As of the latest reported scores on GenAIList, Qwen3-235B-A22B achieves the highest result on MT-AIME 2024 with a score of 80.8.
Is a higher MT-AIME 2024 score better?
Yes. On MT-AIME 2024 a higher score indicates better performance, so models near the top of the leaderboard are the strongest.