muse-wonder: benchmark results
Provider: Meta. Access: Open.
Aggregate Rating 1873 ± 33, rank #22 of 1639 rated models, from 22 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Vals AI CyberBench | 80.36 | Accuracy (%) | 100 |
| Vals AI Code Migration | 62.55 | Accuracy (%) | 93.4 |
| Vals AI Legal Research Bench | 50.48 | Accuracy (%) | 92.9 |
| Vals AI Harvey Legal Agent Bench | 17.08 | Accuracy (%) | 91 |
| Vals AI Public Benefits Bench | 70.23 | Accuracy (%) | 89.8 |
| Gert Labs Rankings | 61.93 | GScore (%) | 86.9 |
| Vals AI Tax Agent Bench | 36.27 | Accuracy (%) | 86.8 |
| Vals AI Vibe Code Bench | 83.09 | Accuracy (%) | 84.3 |
| Vals AI Terminal-Bench 4.0 | 49.49 | Accuracy (%) | 83 |
| Vals AI (Vals Index) | 61.06 | Accuracy (%) | 80.9 |
| Tinybird AI SQL Benchmark - First-Attempt Success Rate | 98 | Questions answered with a valid query on the first attempt ( | 80.4 |
| Vals AI Finance Agent v2 | 56.64 | Accuracy (%) | 79.5 |
Interactive version: theaggregate.ai/model?slug=muse-wonder-h477ot · How It Works · Data refreshed daily, snapshot 2026-10-14.