eeve_dpo-v3: benchmark results
Provider: Yanolja. Released 2024-02-27. Access: Open.
Unified ELO 1558 ± 18, rank #807 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-EQ-Bench | 61.55 | EQ-Bench score | 99.7 |
| Open Korean LLM - Ko-Winogrande | 72.06 | Accuracy (%) | 97.8 |
| Open Korean LLM - KorNAT-CKA | 36.15 | Normalized accuracy (%) | 96.3 |
| Open Korean LLM - KorNAT-Harmless | 71.9 | Normalized accuracy (%) | 95 |
| Open Korean LLM Leaderboard | 47.37 | Average Score (%) | 94.9 |
| Open Ko-LLM Leaderboard | 47.37 | Average (%) | 94.8 |
| Open Korean LLM - KorNAT-Helpful | 50.15 | Normalized accuracy (%) | 84.2 |
| Open Korean LLM - Ko-IFEval | 32.87 | Strict accuracy, prompt/instruction mean (%) | 64.2 |
| Open Korean LLM - Ko-GPQA-Diamond | 24.24 | Normalized accuracy (%) | 62.4 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 21.38 | Exact match, flexible extract (%) | 53.7 |
Interactive version: theaggregate.ai/model?slug=eeve-dpo-v3 · How It Works · Data refreshed daily, snapshot 2026-09-19.