MMAU-Pro - Voice Chat: leaderboard
Audio-language models answer human-written questions about real-world speech, sound and music clips of up to 10 minutes; the score is accuracy on the 290 spoken STEM, prosody and world-knowledge questions, from the authors' runs. Unlike the original MMAU, clips are long, can be mixed or multiple, and include spatial audio, voice chat and open-ended questions.
Metric: Accuracy (%). Source: arxiv.org. Saturation forecast: Estimated already saturated. 18 models tracked. Updated 17 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen2.5-Omni-7B | 60 | #1883 |
| 2 | Audio Flamingo 3 | 58.6 | |
| 3 | Gemma 3n E4B (IT) | 58.3 | #2432 |
| 4 | DeSTA2-8B-beta | 54.8 | |
| 5 | SALMONN-13B | 53.2 | |
| 6 | Gemma 3n E2B (IT) | 51.4 | #4283 |
| 7 | DeSTA2.5-Audio | 51 | #2076 |
| 8 | Qwen2.5-Omni-3B | 46.5 | #3596 |
| 9 | Ming-Lite-Omni-1.5 | 44.5 | #2348 |
| 10 | Audio Reasoner | 43.4 | #3028 |
| 11 | Phi-4 Multimodal Instruct | 42.7 | #4209 |
| 12 | Baichuan-Omni-1.5 | 40 | #3064 |
| 13 | Audio Flamingo 2 | 37.2 | |
| 14 | SALMONN-7B | 36.5 | |
| 15 | R1-AQA | 32.7 | #5707 |
No result here: #1 Claude Fable 5.2, #2 GPT-6.2 Astra, #3 Gemini 4 Carbon.
Interactive version: theaggregate.ai/benchmark?slug=mmau-pro-voice-chat · How It Works · Data refreshed daily, snapshot 2026-11-17.