| Benchmark | Gemini 2.5 Flash | Muse Spark 1.1 |
|---|---|---|
| AA Agentic Index | 18.8 | 39.7 |
| AA Intelligence | 17.7 | 53.2 |
| AA-LCR | 65.7 | 81.3 |
| AA-Omniscience | -29.8 | 28.1 |
| arena_text_factuality | 1437 | 1466 |
| Artificial Analysis Coding Index | 22.2 | 71.3 |
| critpt | 1.4 | 15.1 |
| gdpval | 11.9 | 43.7 |
| GPQA Diamond | 82.8 | 89.8 |
| HLE | 12.1 | 62.1 |
| OmniScience Accuracy | 26.1 | 52 |
| OmniScience Non-Hallucination | 24.7 | 50 |
| scicode | 35.9 | 58.2 |
| VisualToolBench | 4.7 | 44.8 |
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.