Muse Wonder: benchmark results
Meta's Muse Wonder (October 12, 2026), a closed model available through the Meta Model API and Muse Code, with a one-million-token context. Provider: Meta. Access: API.
Aggregate Rating 1913 ± 26, rank #13 of 1640 rated models, from 69 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA GDP.pdf - Legal - Criterion Pass Rate | 88.03 | Criterion Pass Rate (%) | 100 |
| AA-Briefcase - DOCX - Rubric Pass Rate | 65.68 | Rubric Pass Rate (%) | 100 |
| AA GDPval | 1770.34 | ELO | 98.6 |
| Artificial Analysis Intelligence Index | 52.39 | Intelligence Index | 98.6 |
| AutomationBench-AA - HR - Raw Objective Completion | 78.45 | Raw Objective Completion (%) | 98.6 |
| AA-Briefcase - PDF - Rubric Pass Rate | 65.52 | Rubric Pass Rate (%) | 98.2 |
| AA-Briefcase - XLSX - Rubric Pass Rate | 61.14 | Rubric Pass Rate (%) | 98.2 |
| LLM Stats (DeepSWE 1.1) | 78.8 | Score (%) | 97.7 |
| AA-Briefcase | 61.21 | Rubric Pass Rate (%) | 97.6 |
| AutomationBench-AA | 68.09 | Guardrail-Adjusted Objective Completion (%) | 97.3 |
| AutomationBench-AA - Strict | 39.12 | Strict Task Success (%) | 97.3 |
| AA GDP.pdf - STEM/Research - Criterion Pass Rate | 79.78 | Criterion Pass Rate (%) | 97.2 |
Interactive version: theaggregate.ai/model?slug=muse-wonder · How It Works · Data refreshed daily, snapshot 2026-10-13.