| Benchmark | Gemini 3.1 Pro | Nemotron 3 Super 120B A12B |
|---|---|---|
| AA Agentic Index | 23 | 8.8 |
| AA Intelligence | 47.7 | 25.7 |
| AA-LCR | 79 | 60.3 |
| AA-Omniscience | 32.9 | -41.5 |
| aime_2026 | 98.3 | 91.7 |
| apexAgents | 33.5 | 1.8 |
| Artificial Analysis Coding Index | 68.8 | 37.7 |
| browsecomp | 85.9 | 31.3 |
| critpt | 17.7 | 3.1 |
| gdpval | 23.3 | 9.9 |
| GPQA Diamond | 94.3 | 82.7 |
| HLE | 51.4 | 22.8 |
| hmmt_feb_2026 | 94.7 | 85.2 |
| IFBench | 77.1 | 72.6 |
| itbenchSre | 30.3 | 1.1 |
| LiveCodeBench | 91.7 | 82.1 |
| mmlu | 92.6 | 86 |
| MMLU-Pro | 91 | 83.8 |
| multichallenge | 71.4 | 55.2 |
| OmniScience Accuracy | 55.3 | 24.3 |
| OmniScience Non-Hallucination | 50.1 | 13 |
| scicode | 59 | 42 |
| SWE-bench Multilingual | 76.9 | 45.8 |
| SWE-bench Verified | 80.6 | 53.7 |
| TauBench V3 - Banking | 21.4 | 10.3 |
| Terminal-Bench 2.0 | 68.5 | 31 |
| Terminal-Bench 2.1 | 74 | 38.6 |
| Terminal-Bench Hard | 68.5 | 28.8 |
| τ²-Bench Telecom (AA run) | 99.3 | 67.8 |
| τ³-Bench | 67.1 | 10.1 |
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.