Llama 3.1 Nemotron 70B Instruct: benchmark results
NVIDIA's RLHF-tuned Llama 3.1 70B (HelpSteer2-Preference + REINFORCE), topping Arena Hard and AlpacaEval 2 at release (October 2024). Provider: NVIDIA. Released 2024-10-15. Access: Open.
Unified ELO 1526 ± 1, rank #560 of 1392 rated models, from 259 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Portuguese | 68.25 | Average Score (%) | 97.7 |
| Open Japanese LLM - MMLU EN Exact Match | 81.23 | Score (%) | 97.6 |
| Open Japanese LLM - Jsick Exact Match | 86.62 | Score (%) | 97.1 |
| Open CoT - LSAT Logical Reasoning | 21.96 | CoT Gain (%) | 96.9 |
| Open PL LLM - Generative | 69.13 | Average Generative Score (%) | 96.9 |
| StickToYourRole | 80.66 | Cardinal Score | 96.8 |
| Open Japanese LLM - Wikicorpus E TO J Bleu JA | 16.75 | Score (%) | 95.9 |
| Thai LLM NLG - xl_sum_tha_seacrowd_t2t (ROUGE1) | 34.87 | ROUGE1 | 95.5 |
| Enkrypt AI - Bias Risk | 45.74 | Risk Score | 94.7 |
| Thai LLM NLU - belebele_tha_thai_seacrowd_qa | 88.33 | Accuracy (%) | 94.2 |
| EuroEval Finnish NLU - Scandisent FI | 92.66 | Sentiment classification Score (%) | 94.1 |
| Open Japanese LLM - Wiki Coreference SET F1 | 9.84 | Score (%) | 94 |
Interactive version: theaggregate.ai/model?slug=llama-3-1-nemotron-70b-instruct · How It Works · Data refreshed daily, snapshot 2026-09-05.