Nemotron 3 Ultra 550B A55B (Reasoning): benchmark results
Provider: NVIDIA. Released 2026-06-04. Access: Open.
Unified ELO 1599 ± 1, rank #444 of 1761 rated models, from 38 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA IFBench | 81.36 | Accuracy (%) | 99.3 |
| AA Long Context Reasoning | 79.33 | Accuracy (%) | 86.9 |
| AA Terminal-Bench Hard | 36.36 | Accuracy (%) | 83.3 |
| AA Omniscience - Software Engineering (SWE) - TypeScript | 36.67 | Accuracy (%) | 82.4 |
| AA GPQA Diamond | 86.67 | Accuracy (%) | 81.7 |
| AA Omniscience - Software Engineering (SWE) - C | 50 | Accuracy (%) | 81.2 |
| AA Omniscience | -0.4 | Score | 80.7 |
| AA Omniscience - Software Engineering (SWE) - HTML | 44 | Accuracy (%) | 80.7 |
| Artificial Analysis Intelligence Index | 29.55 | Intelligence Index | 80.7 |
| AA Humanity's Last Exam | 28.41 | Accuracy (%) | 80.2 |
| AA CritPt | 3.14 | Accuracy (%) | 74.6 |
| AA Omniscience - Software Engineering (SWE) - Java | 20 | Accuracy (%) | 74.1 |
Interactive version: theaggregate.ai/model?slug=nemotron-3-ultra-550b-a55b-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-05.