AudioBench - MMAU Music: leaderboard

AudioBench music slice using MMAU music questions judged by Llama 3 70B.

Metric: Score (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 10 models tracked.

Top models

#ModelScoreOverall rank
1Phi-4 Multimodal Instruct68.86#906
2SeaLLMs-Audio-7B66.47#1018
3Qwen2-Audio-7B-Instruct60.18#880
4Qwen-Audio-Chat59.58#1244
5Gemini 1.5 Flash58.68#618
6SALMONN-7B55.99#1293

No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.

Interactive version: theaggregate.ai/benchmark?slug=audiobench-mmau-music · How It Works · Data refreshed daily, snapshot 2026-10-20.