Audio Understanding on MMAU Pro (test)
59.2PerformanceGemini 2.5 Flash
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Gemini 2.5 Flash2025.12 | 59.2 | — | |
| MiMo-Audio-7B-Instruct2025.12 | 53.35 | — | |
| GPT-4o-Audio2025.12 | 52.5 | — | |
| Qwen2.5-Omni2025.12 | 52.2 | — | |
| Audio Flamingo 32025.12 | 51.7 | — | |
| Step-Audio2-mini2025.12 | 47.91 | — | |
| Kimi-Audio-Instruct2025.12 | 46.6 | — | |
| GLM-4-Voice2025.12 | 38.25 | — | |
| Audio-FlamingoModel version=32025.12 | — | 51.7 | |
| Fun-Audio-ChatModel size=30B-A3B2025.12 | — | 59.9 | |
| Fun-Audio-ChatModel size=8B2025.12 | — | 58 | |
| Kimi-Audio2025.12 | — | 46.6 | |
| MiMo-Audio2025.12 | — | 53.4 | |
| Step-Audio2-Mini2025.12 | — | 53.2 |