Malpyung_EEVE-Korean-10.8B-v1.0_LoRA: benchmark results
Provider: Other. Access: Open.
Unified ELO 1559 ± 17, rank #804 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-Winogrande | 70.09 | Accuracy (%) | 96.4 |
| Open Korean LLM - Ko-EQ-Bench | 53.37 | EQ-Bench score | 96.2 |
| Open Korean LLM Leaderboard | 47.93 | Average Score (%) | 95.1 |
| Open Ko-LLM Leaderboard | 47.93 | Average (%) | 95 |
| Open Korean LLM - KorNAT-CKA | 34.65 | Normalized accuracy (%) | 94.2 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 49.73 | Exact match, flexible extract (%) | 83.6 |
| Open Korean LLM - KorNAT-Harmless | 65.51 | Normalized accuracy (%) | 80.9 |
| Open Korean LLM - Ko-IFEval | 38.81 | Strict accuracy, prompt/instruction mean (%) | 80.4 |
| Open Korean LLM - KorNAT-Helpful | 48.43 | Normalized accuracy (%) | 67.9 |
| Open Korean LLM - Ko-GPQA-Diamond | 24.24 | Normalized accuracy (%) | 62.4 |
Interactive version: theaggregate.ai/model?slug=malpyung-eeve-korean-10-8b-v1-0-lora · How It Works · Data refreshed daily, snapshot 2026-09-19.