| Benchmark | Gemini 3.6 Flash | Kimi K2 (Reasoning) |
|---|---|---|
| AA Agentic Index | 40.5 | 47.9 |
| AA Intelligence | 52 | 33.5 |
| AA-LCR | 79 | 70.3 |
| AA-Omniscience | 22.1 | -21.4 |
| Artificial Analysis Coding Index | 69.2 | 34.8 |
| critpt | 10.6 | 2.6 |
| gdpval | 46.1 | 24.5 |
| GPQA Diamond | 92.8 | 84.5 |
| HLE | 40.8 | 23.9 |
| OmniScience Accuracy | 50 | 30.9 |
| OmniScience Non-Hallucination | 44.4 | 25.8 |
| scicode | 52.7 | 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.