Llama-3-Instruct-8B-RDPO-v0.2: benchmark results
Princeton NLP's R-DPO (length-regularized DPO) v0.2 baseline on Llama 3 8B Instruct, from the SimPO paper's updated baseline suite. Provider: Princeton NLP. Released 2024-07-06. Access: Open.
Aggregate Rating 1451 ± 19, rank #997 of 1661 rated models, from 21 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-GPQA-Diamond | 27.27 | Normalized accuracy (%) | 92.4 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 57.01 | Exact match, flexible extract (%) | 91.3 |
| Open Korean LLM - KorNAT-Helpful | 51.45 | Normalized accuracy (%) | 90.7 |
| Open Korean LLM - Ko-IFEval | 45.64 | Strict accuracy, prompt/instruction mean (%) | 90.1 |
| Open LLM Leaderboard - IFEval | 70.77 | Score | 84 |
| Open Korean LLM Leaderboard | 41.36 | Average Score (%) | 82.2 |
| Open Ko-LLM Leaderboard | 41.36 | Average (%) | 82.1 |
| Open Korean LLM - Ko-EQ-Bench | 31.93 | EQ-Bench score | 81.1 |
| Open Portuguese LLM - OAB Exams | 50.62 | Accuracy (%) | 77.9 |
| Open Portuguese LLM - BLUEX | 58.28 | Accuracy (%) | 72.6 |
| Open Portuguese LLM - ENEM | 68.51 | Accuracy (%) | 70.3 |
| Open Korean LLM - KorNAT-CKA | 26.71 | Normalized accuracy (%) | 66.3 |
Interactive version: theaggregate.ai/model?slug=llama-3-instruct-8b-rdpo-v0-2 · How It Works · Data refreshed daily, snapshot 2026-10-23.