New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
AIME 2026 benchmark
AI model leaderboard for the AIME 2026 benchmark. Compare how large language models score on AIME 2026, see the full ranking, and understand what this AI benchmark measures. ERNIE 5.1 currently leads with 99.6%. 2026 American Invitational Mathematics Examination problems.
Leaderboard
| # | Model | Organization | Score | Variant | Source |
|---|---|---|---|---|---|
| #1 | ERNIE 5.1 | Baidu | 99.6% | official ↗ | |
| #2 | Gemma 4 12B | Google DeepMind | 77.5% | official ↗ |
Frequently asked questions about AIME 2026
What is the AIME 2026 benchmark?
2026 American Invitational Mathematics Examination problems.
How is the AIME 2026 benchmark scored?
AIME 2026 is scored using the accuracy (%) metric, where a higher score is better. The maximum achievable score is 100.000. GenAIList aggregates reported scores from model providers and papers into a single ranked leaderboard.
Which AI model scores highest on AIME 2026?
As of the latest reported scores on GenAIList, ERNIE 5.1 achieves the highest result on AIME 2026 with a score of 99.6%.
Is a higher AIME 2026 score better?
Yes. On AIME 2026 a higher score indicates better performance, so models near the top of the leaderboard are the strongest.