# MMMLU — AI model rankings

average accuracy on multilingual translations of MMLU

12 tracked models have a published MMMLU 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 | Gemini 3.1 Pro | Google | 92.6% | Lab | Feb 19 2026 |
| 2 | Gemini 3.0 Pro | Google | 91.8% | Lab | Nov 18 2025 |
| 2 | Gemini 3.0 Flash | Google | 91.8% | Lab | Dec 17 2025 |
| 4 | GPT-5.2 | OpenAI | 89.6% | Lab | Dec 11 2025 |
| 5 | Gemini 3.1 Flash-Lite | Google | 88.9% | Lab | Mar 3 2026 |
| 6 | Gemma 4 | Google | 88.4% | Lab | Apr 2 2026 |
| 7 | GPT-4.1 | OpenAI | 87.3% | Lab | Apr 14 2025 |
| 8 | GPT-4.5 | OpenAI | 85.1% | Lab | Feb 27 2025 |
| 9 | gpt-oss-120b | OpenAI | 81.3% | Lab | Aug 5 2025 |
| 10 | GPT-4.1 mini | OpenAI | 78.5% | Lab | Apr 14 2025 |
| 11 | gpt-oss-20b | OpenAI | 75.7% | Lab | Aug 5 2025 |
| 12 | GPT-4.1 nano | OpenAI | 66.9% | Lab | Apr 14 2025 |


---

Canonical page: https://aireleasetracker.com/benchmark/mmmlu
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.