Audio Reasoner: benchmark results
Provider: Other.
Aggregate Rating 1474 ± 39, rank #2769 of 10916 rated models, from 31 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| SonicBench - Brightness | 73 | Accuracy (%; spectral brightness (NSynth); 100 recognition a | 80.6 |
| SonicBench - Duration | 55 | Accuracy (%; sound-event duration (TAU); 100 recognition and | 77.4 |
| MedMosaic - Cough Sound MCQ | 48.1 | Accuracy (%; multiple choice, extracted answer letter agains | 75 |
| SonicBench - Loudness | 56 | Accuracy (%; loudness (integrated LUFS contrasts of at least | 74.2 |
| SonicBench - Pitch | 67 | Accuracy (%; pitch height (NSynth notes, one-semitone contra | 74.2 |
| MedMosaic - Lung Sound MCQ | 41.7 | Accuracy (%; multiple choice, extracted answer letter agains | 66.7 |
| SonicBench | 55 | Accuracy (%; mean over 12 physical attributes of 100 recogni | 59.7 |
| BASS - Artist Duration | 3.06 | Exact-match accuracy (%) on the 327 artist duration question | 53.8 |
| BASS - Gene Dominance Ranking | 6.06 | Exact-match accuracy (%) on the 33 gene dominance ranking qu | 53.8 |
| MMAU | 63.78 | Avg Test (imported) | 50 |
| SonicBench - Texture | 72 | Accuracy (%; sound texture (ESC-50); 100 recognition and 100 | 50 |
| SonicBench - Timbre | 54 | Accuracy (%; instrument timbre (NSynth); 100 recognition and | 50 |
Interactive version: theaggregate.ai/model?slug=audio-reasoner · How It Works · Data refreshed daily, snapshot 2026-11-08.