# MMMU-Pro — AI model rankings

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

9 tracked models have a published MMMU-Pro score. Higher is better. Scores are as published at each model's release.

## Ranking

| Rank | Model | Developer | Score | Released |
| --- | --- | --- | --- | --- |
| 1 | Gemini 3.5 Flash | Google | 83.6% | May 19 2026 |
| 2 | Kimi K3 | Moonshot AI | 81.6% | Jul 16 2026 |
| 3 | Gemini 3.0 Flash | Google | 81.2% | Dec 17 2025 |
| 3 | GPT-5.5 | OpenAI | 81.2% | Apr 23 2026 |
| 5 | Gemini 3.1 Pro | Google | 80.5% | Feb 19 2026 |
| 6 | Qwen3.5 | Qwen | 79% | Feb 16 2026 |
| 7 | Qwen3.6 | Qwen | 75.3% | Apr 16 2026 |
| 8 | Claude Opus 4.7 | Anthropic | 75.2% | Apr 16 2026 |
| 9 | Claude Sonnet 4.6 | Anthropic | 74.5% | Feb 17 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). Benchmark scores are the figures published by the releasing lab at launch.