llama3.1-korean-v1.5-sft-by-aidx: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1544 ± 17, rank #865 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-CKA | 34.49 | Normalized accuracy (%) | 93.9 |
| Open Korean LLM Leaderboard | 46.58 | Average Score (%) | 93.5 |
| Open Ko-LLM Leaderboard | 46.58 | Average (%) | 93.4 |
| Open Korean LLM - Ko-IFEval | 46.41 | Strict accuracy, prompt/instruction mean (%) | 91.2 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 56.79 | Exact match, flexible extract (%) | 90.8 |
| Open Korean LLM - Ko-EQ-Bench | 37.56 | EQ-Bench score | 88.4 |
| Open Korean LLM - Ko-Winogrande | 60.69 | Accuracy (%) | 56.6 |
| Open Korean LLM - KorNAT-Helpful | 46.45 | Normalized accuracy (%) | 46.5 |
| Open Korean LLM - KorNAT-Harmless | 62.29 | Normalized accuracy (%) | 41.3 |
| Open Korean LLM - Ko-GPQA-Diamond | 22.73 | Normalized accuracy (%) | 40.9 |
Interactive version: theaggregate.ai/model?slug=llama3-1-korean-v1-5-sft-by-aidx · How It Works · Data refreshed daily, snapshot 2026-09-19.