# MMMU-Pro — AI model rankings

A tougher version of MMMU — college-level questions that mix images, diagrams, and text together. Higher is better.

12 tracked models have a published MMMU-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 | Gemini 3.5 Flash | Google | 83.6% | Lab | May 19 2026 |
| 2 | Qwen3.8-Max-0902 | Qwen | 82.7% | Lab | Sep 2 2026 |
| 3 | Qwen3.8-Max | Qwen | 82.3% | Lab | Aug 3 2026 |
| 4 | Kimi K3 | Moonshot AI | 81.6% | Lab | Jul 16 2026 |
| 5 | Gemini 3.0 Flash | Google | 81.2% | Lab | Dec 17 2025 |
| 5 | GPT-5.5 | OpenAI | 81.2% | Lab | Apr 23 2026 |
| 7 | Gemini 3.1 Pro | Google | 80.5% | Lab | Feb 19 2026 |
| 8 | Qwen3.5 | Qwen | 79% | Lab | Feb 16 2026 |
| 9 | Qwen3.6 | Qwen | 75.3% | Lab | Apr 16 2026 |
| 10 | Claude Opus 4.7 | Anthropic | 75.2% | Lab | Apr 16 2026 |
| 11 | Claude Sonnet 4.6 | Anthropic | 74.5% | Lab | Feb 17 2026 |
| 12 | Muse Glimmer | Meta | 74% | Lab | Aug 10 2026 |


---

Canonical page: https://aireleasetracker.com/benchmark/mmmu-pro
Full dataset: https://aireleasetracker.com/llms-full.txt · JSON: https://aireleasetracker.com/models.json
Source: AI Release Tracker (https://aireleasetracker.com). Most benchmark scores come from lab launch material; gathered results identify the leaderboard that published them.