Llama-3-Instruct-8B-IPO: benchmark results
Provider: Princeton NLP. Access: Open.
Aggregate Rating 1488 ± 40, rank #2531 of 10916 rated models, from 14 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-EQ-Bench | 45.06 | EQ-Bench score | 93.9 |
| Open Korean LLM - Ko-IFEval | 48.79 | Strict accuracy, prompt/instruction mean (%) | 92.7 |
| Open Korean LLM Leaderboard | 43.16 | Average Score (%) | 87.5 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 53.68 | Exact match, flexible extract (%) | 87.3 |
| Open Korean LLM - Ko-GPQA-Diamond | 25.76 | Normalized accuracy (%) | 81.3 |
| Open Portuguese LLM - ASSIN2 RTE | 91.95 | Macro F1 (%) | 72.8 |
| Open Portuguese LLM - ENEM | 68.51 | Accuracy (%) | 70.3 |
| Open Portuguese LLM - OAB Exams | 48.34 | Accuracy (%) | 69.4 |
| Open Portuguese LLM - BLUEX | 55.49 | Accuracy (%) | 63.4 |
| Open Korean LLM - KorNAT-CKA | 26.12 | Normalized accuracy (%) | 62.3 |
| Open Korean LLM - Ko-Winogrande | 61.09 | Accuracy (%) | 59.7 |
| Open Portuguese LLM - FaQuAD NLI | 68.85 | Macro F1 (%) | 57.7 |
Interactive version: theaggregate.ai/model?slug=llama-3-instruct-8b-ipo · How It Works · Data refreshed daily, snapshot 2026-11-08.