Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) — benchmark results
Llama 3.3 Nemotron Super 49B v1 evaluated with reasoning disabled. Provider: NVIDIA. Released 2025-03-18. Access: Open.
Unified ELO 1460 ± 32, rank #956 of 1776 rated models, from 36 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA MATH-500 | 77.53 | Accuracy (%) | 43.1 |
| AA Omniscience - Software Engineering (SWE) - PHP | 20 | Accuracy (%) | 40.8 |
| AA Omniscience | -49.68 | Score | 38.8 |
| AA IFBench | 39.46 | Accuracy (%) | 37.1 |
| AA MMLU-Pro | 69.77 | Accuracy (%) | 36 |
| AA Omniscience - Software Engineering (SWE) - Java | 14 | Accuracy (%) | 35.2 |
| AA Omniscience - Software Engineering (SWE) - HTML | 26 | Accuracy (%) | 34.9 |
| AA Omniscience - Software Engineering (SWE) - C | 28 | Accuracy (%) | 34.4 |
| AA Omniscience - Software Engineering (SWE) - Go | 12 | Accuracy (%) | 32.6 |
| AA Omniscience - Software Engineering (SWE) - Rust | 42 | Accuracy (%) | 32.4 |
| AA Omniscience - Health | 15.2 | Accuracy (%) | 32.3 |
| AA LiveCodeBench | 28.04 | Pass@1 (%) | 31.4 |
Interactive version: theaggregate.ai/model?slug=llama-3-3-nemotron-super-49b-v1-non-reasoning · How the rankings work · Data refreshed daily, snapshot 2026-07-22.