koko_test: benchmark results
Provider: Other. Access: Open.
Unified ELO 1266 ± 18, rank #2547 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-GPQA-Diamond | 28.28 | Normalized accuracy (%) | 96.6 |
| Open Korean LLM - KorNAT-CKA | 28.45 | Normalized accuracy (%) | 75.6 |
| Open Korean LLM - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
| Open Korean LLM - Ko-IFEval | 15.1 | Strict accuracy, prompt/instruction mean (%) | 10.4 |
| Open Korean LLM - KorNAT-Harmless | 59.71 | Normalized accuracy (%) | 8.6 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 0 | Exact match, flexible extract (%) | 2 |
| Open Korean LLM Leaderboard | 27.35 | Average Score (%) | 1.9 |
| Open Ko-LLM Leaderboard | 27.35 | Average (%) | 1.8 |
| Open Korean LLM - Ko-Winogrande | 49.33 | Accuracy (%) | 1.6 |
| Open Korean LLM - KorNAT-Helpful | 35.93 | Normalized accuracy (%) | 1.6 |
Interactive version: theaggregate.ai/model?slug=koko-test · How It Works · Data refreshed daily, snapshot 2026-09-19.