MMAU: leaderboard
MMAU evaluates audio understanding across sound, music, and speech tasks with test-mini and test splits.
Metric: Avg Test (imported). Source: sakshi113.github.io. Source: benchmarklist.com. Saturation forecast: Around September 2027. 31 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) (High) | 82.5 | #58 (Gemini 3.1 Pro (Preview)) |
| 2 | Inkling | 77.2 | #126 |
| 3 | Inkling Small | 77 | #138 |
| 4 | Gemini 2.5 Pro | 69.36 | #152 |
| 5 | Gemini 2.5 Flash | 67.39 | #245 |
| 6 | Gemini 2.0 Flash | 67.03 | #342 |
| 7 | Gemini 2.5 Flash Lite | 61.61 | #422 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=mmau · How It Works · Data refreshed daily, snapshot 2026-10-20.