Llama-3-Base-8B-SFT-SLiC-HF: benchmark results
Princeton NLP's SLiC-HF baseline on its Llama 3 8B SFT base, part of the SimPO paper's preference-optimization baseline suite. Provider: Princeton NLP. Released 2024-07-06. Access: Open.
Aggregate Rating 1456 ± 19, rank #963 of 1661 rated models, from 21 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-EQ-Bench | 39.87 | EQ-Bench score | 91.3 |
| Open Korean LLM Leaderboard | 43.78 | Average Score (%) | 88.8 |
| Open Ko-LLM Leaderboard | 43.78 | Average (%) | 88.6 |
| Open Korean LLM - KorNAT-Helpful | 50.95 | Normalized accuracy (%) | 88.1 |
| Open Korean LLM - KorNAT-Harmless | 65.39 | Normalized accuracy (%) | 79.9 |
| Open Korean LLM - Ko-IFEval | 37.74 | Strict accuracy, prompt/instruction mean (%) | 79 |
| Open Korean LLM - Ko-GPQA-Diamond | 25.25 | Normalized accuracy (%) | 75.9 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 36.09 | Exact match, flexible extract (%) | 69.4 |
| Open Korean LLM - Ko-Winogrande | 61.88 | Accuracy (%) | 64.9 |
| Open Portuguese LLM - OAB Exams | 46.92 | Accuracy (%) | 64.8 |
| Open Portuguese LLM - ENEM | 65.92 | Accuracy (%) | 63 |
| Open LLM Leaderboard - IFEval | 48.9 | Score | 57.1 |
Interactive version: theaggregate.ai/model?slug=llama-3-base-8b-sft-slic-hf · How It Works · Data refreshed daily, snapshot 2026-10-23.