Across 26 shared benchmarks, Nemotron 3 Super 120B A12B scores higher on 4 and Qwen3.5 122B A10B on 21, with 1 level. The widest gap is critpt, where Nemotron 3 Super 120B A12B scores 3.1 against 0.9. Nemotron 3 Super 120B A12B is the cheaper of the two on tracked API pricing ($0.20 against $0.40 per million input tokens).
| Benchmark | Nemotron 3 Super 120B A12B | Qwen3.5 122B A10B |
|---|---|---|
| AA Agentic Index | 8.8 | 21.3 |
| AA Intelligence | 25.7 | 32.8 |
| AA-LCR | 60.3 | 70.3 |
| AA-Omniscience | -41.5 | -41.5 |
| Artificial Analysis Coding Index | 37.7 | 45.7 |
| browsecomp | 31.3 | 63.8 |
| critpt | 3.1 | 0.9 |
| gdpval | 9.9 | 24.3 |
| GPQA Diamond | 82.7 | 86.6 |
| GSM8K | 90.7 | 94.5 |
| HLE | 22.8 | 47.5 |
| HMMT 2025 | 94.7 | 90.3 |
| IFBench | 72.6 | 76.1 |
| mmlu_prox | 79.4 | 82.2 |
| MMLU-Pro | 83.8 | 86.7 |
| multichallenge | 55.2 | 61.5 |
| OmniScience Accuracy | 24.3 | 24.4 |
| OmniScience Non-Hallucination | 13 | 12.9 |
| scicode | 42 | 42 |
| SWE-bench Verified | 53.7 | 72 |
| TauBench V3 - Banking | 10.3 | 15.3 |
| Terminal-Bench 2.0 | 31 | 49.4 |
| Terminal-Bench 2.1 | 38.6 | 47.6 |
| Terminal-Bench Hard | 28.8 | 31.1 |
| τ²-Bench Telecom (AA run) | 67.8 | 93.6 |
| τ³-Bench | 10.1 | 13.6 |
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. Quoted rates are the price-setter row we currently track for each model — its direct or vendor-official listing where one exists (direct, direct), otherwise the lowest tracked offer.