# AIME 2025 — AI model rankings

single-answer accuracy on the 2025 invitational mathematics exam without tools

12 tracked models have a published AIME 2025 score. Higher is better. Scores come from published lab reports and benchmark sources; recorded sources appear beside the scores.

## Ranking

| Rank | Model | Developer | Score | Source | Released |
| --- | --- | --- | --- | --- | --- |
| 1 | GPT-5.2 | OpenAI | 100% | Lab | Dec 11 2025 |
| 2 | Gemini 3.0 Flash | Google | 95.2% | Lab | Dec 17 2025 |
| 3 | Gemini 3.0 Pro | Google | 95% | Lab | Nov 18 2025 |
| 4 | GPT-5 | OpenAI | 94.6% | Lab | Aug 7 2025 |
| 5 | GPT-5.1 | OpenAI | 94% | Lab | Nov 12 2025 |
| 6 | gpt-oss-120b | OpenAI | 92.5% | Lab | Aug 5 2025 |
| 7 | gpt-oss-20b | OpenAI | 91.7% | Lab | Aug 5 2025 |
| 8 | GPT-5 mini | OpenAI | 91.1% | Lab | Aug 7 2025 |
| 9 | Gemini 2.5 Pro | Google | 86.7% | Lab | Mar 25 2025 |
| 10 | GPT-5 nano | OpenAI | 85.2% | Lab | Aug 7 2025 |
| 11 | Gemini 2.5 Flash | Google | 78% | Lab | Apr 17 2025 |
| 12 | Gemini 2.5 Flash-Lite | Google | 49.8% | Lab | Jun 17 2025 |


---

Canonical page: https://aireleasetracker.com/benchmark/aime-2025
Site index: https://aireleasetracker.com/llms.txt
Source: AI Release Tracker (https://aireleasetracker.com). Most benchmark scores come from lab launch material; gathered results identify the leaderboard that published them.