Muse Glimmer 30B (Medium): benchmark results
Provider: Meta. Released 2026-08-09. Access: Open.
Aggregate Rating 1623 ± 12, rank #630 of 2167 rated models, from 65 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Meta Glimmer Reasoning Report - Omnilingual-GAIA2 | 60.7 | Published Score (%) | 100 |
| Meta Glimmer Reasoning Report - PinchBench | 90.9 | Published Score (%) | 100 |
| Meta Glimmer Reasoning Report - TauBench (Retail) | 88.4 | Published Score (%) | 100 |
| Medmarks - M-ARC | 86 | Score (%) | 97.7 |
| Medmarks - Med-HALT Reasoning FCT | 90.74 | Score (%) | 93.2 |
| Medmarks - MedConceptsQA Hard | 81.62 | Score (%) | 90.9 |
| Medmarks - MedConceptsQA Medium | 86.73 | Score (%) | 90.9 |
| Medmarks - Med-HALT Reasoning NOTA | 69.31 | Score (%) | 88.6 |
| Medmarks - MedConceptsQA Easy | 99.05 | Score (%) | 85.2 |
| Medmarks - SCTpublic | 71.49 | Score (%) | 84.1 |
| UGI - Natural Intelligence | 36.4 | NatInt Score | 83 |
| Medmarks - LongHealth Task 1 | 89.83 | Score (%) | 79.5 |
Interactive version: theaggregate.ai/model?slug=muse-glimmer-30b-medium · How It Works · Data refreshed daily, snapshot 2026-10-22.