Muse Spark Open (xHigh): benchmark results
Provider: Meta. Released 2026-10-15. Access: Open.
Aggregate Rating 1845 ± 18, rank #76 of 2145 rated models, from 67 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Meta Muse Spark Open Card - BabyVision (Tools, Mean@5) | 80.2 | Mean accuracy (%) | 100 |
| Meta Muse Spark Open Card - IntPhys2 (1012 videos) | 68.3 | Accuracy (%) | 100 |
| Meta Muse Spark Open Card - OSWorld Verified (361 tasks, GUI, mean@4) | 80.4 | Mean task reward (%) | 100 |
| Meta Muse Spark Open Card - SimpleVQA (Tools, Mean@5) | 73.6 | Mean accuracy (%) | 100 |
| Meta Muse Spark Open Card - WorldVQA (Tools, 3000 questions) | 49.4 | Accuracy (%) | 100 |
| Meta Muse Spark Open Card - ZeroBench (Tools, Pass@5) | 50 | Pass@5 (%) | 100 |
| AA GDP.pdf - Legal - Criterion Pass Rate | 87.81 | Criterion Pass Rate (%) | 99.5 |
| AA-Omniscience Hallucination Rate | 24.07 | Hallucination Rate (%) | 96.5 |
| AA GDP.pdf - HR - Criterion Pass Rate | 85.57 | Criterion Pass Rate (%) | 95.9 |
| AA-Omniscience Index - Science, Engineering & Mathematics | 25.8 | Omniscience Index | 95.7 |
| AA-Omniscience Index - Software Engineering (SWE) - Rust | 78 | Omniscience Index | 94.5 |
| AA-Omniscience Index - Health | 13.2 | Omniscience Index | 93.8 |
Interactive version: theaggregate.ai/model?slug=muse-spark-open-xhigh · How It Works · Data refreshed daily, snapshot 2026-10-16.