| Benchmark | GLM-5.1 | Nemotron 3 Super 120B A12B |
|---|---|---|
| AA Agentic Index | 66 | 8.8 |
| AA Intelligence | 41 | 25.7 |
| AA-LCR | 68 | 60.3 |
| AA-Omniscience | 0.8 | -41.5 |
| aime_2026 | 95.8 | 91.7 |
| Artificial Analysis Coding Index | 55.8 | 37.7 |
| browsecomp | 79.3 | 31.3 |
| critpt | 4.6 | 3.1 |
| gdpval | 49.5 | 9.9 |
| GPQA Diamond | 86.8 | 82.7 |
| HLE | 52.3 | 22.8 |
| HMMT 2025 | 94 | 94.7 |
| hmmt_feb_2026 | 89.4 | 85.2 |
| IFBench | 76.3 | 72.6 |
| itbenchSre | 40.3 | 1.1 |
| MMLU-Pro | 86 | 83.8 |
| OmniScience Accuracy | 25.2 | 24.3 |
| OmniScience Non-Hallucination | 70.1 | 13 |
| scicode | 43.8 | 42 |
| SWE-bench Multilingual | 73.3 | 45.8 |
| TauBench V3 - Banking | 13.6 | 10.3 |
| Terminal-Bench 2.0 | 69 | 31 |
| Terminal-Bench 2.1 | 63.5 | 38.6 |
| Terminal-Bench Hard | 43.2 | 28.8 |
| τ²-Bench Telecom (AA run) | 97.7 | 67.8 |
| τ³-Bench | 70.6 | 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.