Llama-3-Instruct-8B-SimPO-v0.2: benchmark results
Princeton NLP's SimPO v0.2 fine-tune of Llama 3 8B Instruct, the updated release from the SimPO reference-free preference optimization paper. Provider: Princeton NLP. Released 2024-07-06. Access: Open.
Aggregate Rating 1455 ± 17, rank #969 of 1656 rated models, from 22 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-Helpful | 55.95 | Normalized accuracy (%) | 96.1 |
| Open Korean LLM - Ko-GPQA-Diamond | 27.27 | Normalized accuracy (%) | 92.4 |
| Open Korean LLM - KorNAT-Harmless | 67.53 | Normalized accuracy (%) | 89.9 |
| Open Korean LLM - Ko-IFEval | 40.86 | Strict accuracy, prompt/instruction mean (%) | 83.8 |
| Open Korean LLM Leaderboard | 41.32 | Average Score (%) | 82 |
| Open Ko-LLM Leaderboard | 41.32 | Average (%) | 81.9 |
| Open LLM Leaderboard - IFEval | 68.09 | Score | 80.8 |
| Open Korean LLM - KorNAT-CKA | 29.24 | Normalized accuracy (%) | 79.1 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 40.18 | Exact match, flexible extract (%) | 74.7 |
| Open LLM Leaderboard - GPQA | 31.46 | Score | 71.4 |
| Open Korean LLM - Ko-Winogrande | 62.12 | Accuracy (%) | 66.6 |
| Open Portuguese LLM - OAB Exams | 47.33 | Accuracy (%) | 65.8 |
Interactive version: theaggregate.ai/model?slug=llama-3-instruct-8b-simpo-v0-2 · How It Works · Data refreshed daily, snapshot 2026-10-20.