ko-Meta-Llama-3.1-8B-Instruct: benchmark results
Provider: Other. Access: Open.
Unified ELO 1458 ± 17, rank #1473 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-GSM8K (flexible extract) | 54.21 | Exact match, flexible extract (%) | 88.2 |
| Open Korean LLM - Ko-IFEval | 40.1 | Strict accuracy, prompt/instruction mean (%) | 81.8 |
| Open Korean LLM - Ko-EQ-Bench | 27.44 | EQ-Bench score | 74.6 |
| Open Korean LLM - KorNAT-Harmless | 64.35 | Normalized accuracy (%) | 73.2 |
| Open Korean LLM Leaderboard | 38.19 | Average Score (%) | 65.3 |
| Open Ko-LLM Leaderboard | 38.19 | Average (%) | 65.2 |
| Open Korean LLM - Ko-GPQA-Diamond | 23.74 | Normalized accuracy (%) | 55.6 |
| Open Korean LLM - KorNAT-Helpful | 46.08 | Normalized accuracy (%) | 42.1 |
| Open Korean LLM - KorNAT-CKA | 21.92 | Normalized accuracy (%) | 37.4 |
| Open Korean LLM - Ko-Winogrande | 56.83 | Accuracy (%) | 28 |
Interactive version: theaggregate.ai/model?slug=ko-meta-llama-3-1-8b-instruct · How It Works · Data refreshed daily, snapshot 2026-09-19.