| Benchmark | Gemini 3.6 Flash | Nemotron 3 Super 120B A12B |
|---|---|---|
| AA Agentic Index | 40.5 | 8.8 |
| AA Intelligence | 52 | 25.7 |
| AA-LCR | 79 | 60.3 |
| AA-Omniscience | 22.1 | -41.5 |
| aime_2026 | 96.7 | 91.7 |
| Artificial Analysis Coding Index | 69.2 | 37.7 |
| critpt | 10.6 | 3.1 |
| gdpval | 46.1 | 9.9 |
| GPQA Diamond | 92.8 | 82.7 |
| HLE | 40.8 | 22.8 |
| hmmt_feb_2026 | 89.4 | 85.2 |
| OmniScience Accuracy | 50 | 24.3 |
| OmniScience Non-Hallucination | 44.4 | 13 |
| scicode | 52.7 | 42 |
| TauBench V3 - Banking | 29.9 | 10.3 |
| Terminal-Bench 2.1 | 78 | 38.6 |
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.