AIME 2024
Created by Mathematical Association of America, 7 Feb 2024 · Higher is better · Website
- OpenAI
- Reported in a later post or card
Results
| Model | Score | Source | Source check |
|---|---|---|---|
| gpt-oss-120b5 Aug 2025 · OpenAI | 96.6% | Introducing gpt-osshigh reasoning effort; tools | |
| gpt-oss-120b5 Aug 2025 · OpenAI | 75.4% | gpt-oss-120b & gpt-oss-20b Model Cardlow reasoning effort; tools; another report, not charted | |
| gpt-oss-120b5 Aug 2025 · OpenAI | 87.9% | gpt-oss-120b & gpt-oss-20b Model Cardmedium reasoning effort; tools; another report, not charted | |
| gpt-oss-20b5 Aug 2025 · OpenAI | 96% | Introducing gpt-osshigh reasoning effort; tools | |
| gpt-oss-20b5 Aug 2025 · OpenAI | 61.2% | gpt-oss-120b & gpt-oss-20b Model Cardlow reasoning effort; tools; another report, not charted | |
| gpt-oss-20b5 Aug 2025 · OpenAI | 86.0% | gpt-oss-120b & gpt-oss-20b Model Cardmedium reasoning effort; tools; another report, not charted | |
| OpenAI o316 Apr 2025 · OpenAI | 95.2% | Introducing gpt-osstools; reported later | |
| o4-mini16 Apr 2025 · OpenAI | 98.7% | Introducing gpt-osstools; reported later |