| Benchmark | GPT-4o mini | Muse Spark 1.1 |
|---|---|---|
| AA Agentic Index | 1 | 39.7 |
| AA Intelligence | 7 | 53.2 |
| Artificial Analysis Coding Index | 11.4 | 71.3 |
| Fortress | 48.1 | 12.4 |
| gdpval | 0 | 43.7 |
| GPQA Diamond | 42.6 | 89.8 |
| HLE | 4.2 | 62.1 |
| scicode | 22.9 | 58.2 |
| TauBench V3 - Banking | 2.9 | 31.8 |
| Terminal-Bench 2.1 | 5.6 | 80 |
Best tracked score per model per benchmark (default configuration; source-attributed). ↓ marks lower-is-better metrics. Open either model for its full surface, provenance and pricing.