| Benchmark | Kimi K2 (Reasoning) | Muse Spark 1.1 |
|---|---|---|
| AA Agentic Index | 47.9 | 39.7 |
| AA Intelligence | 33.5 | 53.2 |
| AA-LCR | 70.3 | 81.3 |
| AA-Omniscience | -21.4 | 28.1 |
| Artificial Analysis Coding Index | 34.8 | 71.3 |
| critpt | 2.6 | 15.1 |
| gdpval | 24.5 | 43.7 |
| GPQA Diamond | 84.5 | 89.8 |
| HLE | 23.9 | 62.1 |
| multichallenge | 66.4 | 75.3 |
| OmniScience Accuracy | 30.9 | 52 |
| OmniScience Non-Hallucination | 25.8 | 50 |
| scicode | 44.8 | 58.2 |
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.