COKAL-DPO_test-v2: benchmark results
Provider: Other. Access: Open.
Unified ELO 1455 ± 17, rank #1494 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-CKA | 40 | Normalized accuracy (%) | 99.2 |
| Open Korean LLM - Ko-Winogrande | 71.67 | Accuracy (%) | 97.7 |
| Open Korean LLM - KorNAT-Harmless | 70.32 | Normalized accuracy (%) | 93.4 |
| Open Korean LLM - KorNAT-Helpful | 51.5 | Normalized accuracy (%) | 90.9 |
| Open Korean LLM Leaderboard | 38.14 | Average Score (%) | 65 |
| Open Ko-LLM Leaderboard | 38.14 | Average (%) | 64.9 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 17.66 | Exact match, flexible extract (%) | 48.6 |
| Open Korean LLM - Ko-IFEval | 22.95 | Strict accuracy, prompt/instruction mean (%) | 25.2 |
| Open Korean LLM - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
| Open Korean LLM - Ko-GPQA-Diamond | 21.21 | Normalized accuracy (%) | 24 |
Interactive version: theaggregate.ai/model?slug=cokal-dpo-test-v2 · How It Works · Data refreshed daily, snapshot 2026-09-19.