# MMLU-Pro — AI model rankings

accuracy on the ten-option MMLU-Pro benchmark

11 tracked models have a published MMLU-Pro 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 | Gemma 4 | Google | 85.2% | Lab | Apr 2 2026 |
| 2 | LLaMA 4 Maverick | Meta | 80.5% | Lab | Apr 5 2025 |
| 3 | Gemini 2.0 Pro | Google | 79.1% | Lab | Feb 5 2025 |
| 4 | Gemini 2.0 Flash | Google | 76.4% | Lab | Dec 11 2024 |
| 5 | Gemini 1.5 Pro-002 | Google | 75.8% | Lab | Sep 24 2024 |
| 6 | LLaMA 4 Scout | Meta | 74.3% | Lab | Apr 5 2025 |
| 7 | LLaMA 3.1 | Meta | 73.3% | Lab | Jul 23 2024 |
| 8 | Gemini 2.0 Flash-Lite | Google | 71.6% | Lab | Feb 5 2025 |
| 9 | LLaMA 3.3 | Meta | 68.9% | Lab | Dec 6 2024 |
| 10 | Gemma 3 | Google | 67.5% | Lab | Mar 12 2025 |
| 11 | Gemini 1.5 Flash-002 | Google | 67.3% | Lab | Sep 24 2024 |


---

Canonical page: https://aireleasetracker.com/benchmark/mmlu-pro
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.