List your company

Audio-Language Understanding on MMAU 1.0 (test)

77.4AccuracyQwen3-Omni (I)
20.82435.51250.264.888Apr 27, 2026

Evaluation results

MethodLinks
Qwen3-Omni (I)Size=30B†, Evaluation Protocol=Full2026.0477.473.1
Qwen3-Omni (T)Size=30B†, Evaluation Protocol=Full2026.0476.273.2
Audio-Flamingo-3Size=8.4B, Evaluation Protocol=Full2026.047581.2
Qwen2.5-Omni-7BSize=10.7B, Evaluation Protocol=Full2026.0474.865.1
DeSTA-2.5Size=8.8B, Evaluation Protocol=Full2026.0465.243.1
Qwen2-Audio-7BSize=8.2B, Evaluation Protocol=Full2026.0463.959.9
Audio-Flamingo-3Size=8.4B, Evaluation Protocol=None2026.0460.9—
Phi-4-MultimodalSize=5.6B, Evaluation Protocol=Full2026.0460.448
Qwen3-Omni (I)Size=30B†, Evaluation Protocol=None2026.0456.6—
Voxtral-Mini-3BSize=4.7B, Evaluation Protocol=Full2026.0455.970.8
Qwen3-Omni (T)Size=30B†, Evaluation Protocol=None2026.0455.8—
Qwen3-Omni (I)Size=30B†, Evaluation Protocol=TB2026.0450.8—
Qwen2.5-Omni-7BSize=10.7B, Evaluation Protocol=None2026.0448.7—
Audio-Flamingo-3Size=8.4B, Evaluation Protocol=TB2026.0445.5—
Qwen2.5-Omni-7BSize=10.7B, Evaluation Protocol=TB2026.0445.5—
Voxtral-Mini-3BSize=4.7B, Evaluation Protocol=None2026.0439.6—
Qwen3-Omni (T)Size=30B†, Evaluation Protocol=TB2026.0438.6—
Qwen2-Audio-7BSize=8.2B, Evaluation Protocol=TB2026.0438.5—
Qwen2-Audio-7BSize=8.2B, Evaluation Protocol=None2026.0438.3—
Phi-4-MultimodalSize=5.6B, Evaluation Protocol=None2026.0429—
Phi-4-MultimodalSize=5.6B, Evaluation Protocol=TB2026.0428.9—
DeSTA-2.5Size=8.8B, Evaluation Protocol=TB2026.0428.4—
DeSTA-2.5Size=8.8B, Evaluation Protocol=None2026.0428.1—
Chance LevelEvaluation Protocol=Full2026.0425—
Chance LevelEvaluation Protocol=None2026.0425—
Chance LevelEvaluation Protocol=TB2026.0425—
Voxtral-Mini-3BSize=4.7B, Evaluation Protocol=TB2026.0423—