Muse Spark 1.2 (xHigh) — benchmark results
Provider: Meta. Released 2026-08-05. Access: API.
Unified ELO 1933 ± 17, rank #41 of 1803 rated models, from 54 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA Long Context Reasoning | 83.33 | Accuracy (%) | 100 |
| Vals AI Finance Agent v2 | 60.6 | Accuracy (%) | 100 |
| Vals AI Harvey Legal Agent Bench | 25.42 | Accuracy (%) | 100 |
| Vals AI TaxEval v2 | 80.38 | Accuracy (%) | 100 |
| Chatbot Arena (Text) | 1498 | Elo | 99.2 |
| Vals AI MedScribe | 90.06 | Accuracy (%) | 98.7 |
| Epoch AI - Scicode | 56.37 | Score | 98.6 |
| Artificial Analysis Intelligence Index | 56.76 | Intelligence Index | 98 |
| AA Omniscience | 27.2 | Score | 97.8 |
| AA Humanity's Last Exam | 45.46 | Accuracy (%) | 97.7 |
| Vals AI CorpFin v2 | 70.94 | Accuracy (%) | 96.9 |
| AA Omniscience - Software Engineering (SWE) - Rust | 84 | Accuracy (%) | 96.4 |
Interactive version: theaggregate.ai/model?slug=muse-spark-1-2-xhigh · How It Works · Data refreshed daily, snapshot 2026-08-08.