ko-Llama-3-8B-sft-v0.3: benchmark results
Provider: Other. Access: Open.
Unified ELO 1454 ± 17, rank #1500 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-IFEval | 38.24 | Strict accuracy, prompt/instruction mean (%) | 79.8 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 37.3 | Exact match, flexible extract (%) | 71.2 |
| Open Korean LLM - Ko-GPQA-Diamond | 24.75 | Normalized accuracy (%) | 69.5 |
| Open Korean LLM Leaderboard | 38.31 | Average Score (%) | 66.1 |
| Open Ko-LLM Leaderboard | 38.31 | Average (%) | 66 |
| Open Korean LLM - KorNAT-Helpful | 48.1 | Normalized accuracy (%) | 64.6 |
| Open Korean LLM - KorNAT-CKA | 24.67 | Normalized accuracy (%) | 52.7 |
| Open Korean LLM - Ko-EQ-Bench | 4.58 | EQ-Bench score | 50.6 |
| Open Korean LLM - Ko-Winogrande | 59.12 | Accuracy (%) | 44.1 |
| Open Korean LLM - KorNAT-Harmless | 61.44 | Normalized accuracy (%) | 28.6 |
Interactive version: theaggregate.ai/model?slug=ko-llama-3-8b-sft-v0-3 · How It Works · Data refreshed daily, snapshot 2026-09-19.