llama3.0-korean-v1.0-sft: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1473 ± 17, rank #1353 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-EQ-Bench | 44.11 | EQ-Bench score | 93.3 |
| Open Korean LLM - Ko-IFEval | 36.92 | Strict accuracy, prompt/instruction mean (%) | 76.9 |
| Open Korean LLM Leaderboard | 40.06 | Average Score (%) | 75.3 |
| Open Ko-LLM Leaderboard | 40.06 | Average (%) | 75.2 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 30.17 | Exact match, flexible extract (%) | 64.8 |
| Open Korean LLM - Ko-GPQA-Diamond | 23.23 | Normalized accuracy (%) | 48.1 |
| Open Korean LLM - Ko-Winogrande | 58.96 | Accuracy (%) | 43 |
| Open Korean LLM - KorNAT-Harmless | 62.1 | Normalized accuracy (%) | 37.9 |
| Open Korean LLM - KorNAT-Helpful | 44.62 | Normalized accuracy (%) | 31.4 |
| Open Korean LLM - KorNAT-CKA | 19.81 | Normalized accuracy (%) | 26.8 |
Interactive version: theaggregate.ai/model?slug=llama3-0-korean-v1-0-sft · How It Works · Data refreshed daily, snapshot 2026-09-19.