GAMA: benchmark results
Provider: Other.
Aggregate Rating 1321 ± 46, rank #8292 of 10556 rated models, from 29 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMAU-Pro - Sound | 45.4 | Accuracy (%) | 76.5 |
| MMAU-Pro - Multi-Audio | 20.2 | Accuracy (%) | 61.8 |
| MedMosaic - Heart Sound MCQ | 36.6 | Accuracy (%; multiple choice, extracted answer letter agains | 45.8 |
| MedMosaic - Lung Sound MCQ | 39.1 | Accuracy (%; multiple choice, extracted answer letter agains | 41.7 |
| SonicBench - Texture | 62 | Accuracy (%; sound texture (ESC-50); 100 recognition and 100 | 40.6 |
| Afrispeech Semantics - Entailment (AfriSpeech-Medical) | 30.3 | Macro-F1 (%; three-way entailment, neutral or contradiction | 40 |
| MedMosaic - Cough Sound MCQ | 33.6 | Accuracy (%; multiple choice, extracted answer letter agains | 33.3 |
| MedMosaic - Multi-Turn MCQ | 35 | Accuracy (%; multiple choice, extracted answer letter agains | 25 |
| MMAU-Pro - Music | 41.2 | Accuracy (%) | 23.5 |
| MMAU-Pro - Instruction Following | 31.7 | Accuracy (%) | 17.6 |
| Afrispeech Semantics - Entailment (AfriSpeech-200) | 19.36 | Macro-F1 (%; three-way entailment, neutral or contradiction | 16.7 |
| Afrispeech Semantics - Plausibility (AfriSpeech-200) | 44.69 | Macro-F1 (%; plausible-but-unsupported versus implausible ju | 16.7 |
Interactive version: theaggregate.ai/model?slug=gama · How It Works · Data refreshed daily, snapshot 2026-11-17.