Llama-3-Instruct-8B-RRHF: benchmark results
Princeton NLP's Llama-3-Instruct-8B-RRHF, a Llama 3 8B Instruct model aligned via RRHF, released alongside Princeton's SimPO preference-optimization paper. Provider: Princeton NLP. Released 2024-07-06. Access: Open.
Aggregate Rating 1434 ± 21, rank #1077 of 1654 rated models, from 20 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-GSM8K (flexible extract) | 54.74 | Exact match, flexible extract (%) | 88.6 |
| Open Korean LLM - Ko-IFEval | 42.72 | Strict accuracy, prompt/instruction mean (%) | 86.8 |
| Open LLM Leaderboard - IFEval | 72.75 | Score | 86.7 |
| Open Korean LLM - Ko-EQ-Bench | 35.75 | EQ-Bench score | 86 |
| Open Portuguese LLM - OAB Exams | 50.66 | Accuracy (%) | 78.1 |
| Open Korean LLM Leaderboard | 39.96 | Average Score (%) | 74.5 |
| Open Korean LLM - KorNAT-Helpful | 48.85 | Normalized accuracy (%) | 73.6 |
| Open Portuguese LLM - ENEM | 69.21 | Accuracy (%) | 72.6 |
| Open Portuguese LLM - BLUEX | 57.3 | Accuracy (%) | 69.4 |
| Open Portuguese LLM - ASSIN2 RTE | 91.03 | Macro F1 (%) | 62.7 |
| Open Korean LLM - Ko-GPQA-Diamond | 24.24 | Normalized accuracy (%) | 62.4 |
| Open Portuguese LLM - FaQuAD NLI | 69.49 | Macro F1 (%) | 59.4 |
Interactive version: theaggregate.ai/model?slug=llama-3-instruct-8b-rrhf · How It Works · Data refreshed daily, snapshot 2026-10-24.