Custom-KoLLM-13B-v4: benchmark results
Provider: Other. Access: Open.
Unified ELO 1368 ± 17, rank #2125 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-Winogrande | 62.98 | Accuracy (%) | 71.6 |
| Open Korean LLM - KorNAT-Helpful | 47.4 | Normalized accuracy (%) | 56.3 |
| Open Korean LLM - KorNAT-Harmless | 62.16 | Normalized accuracy (%) | 39.4 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 12.05 | Exact match, flexible extract (%) | 38.5 |
| Open Korean LLM - KorNAT-CKA | 20.51 | Normalized accuracy (%) | 30.2 |
| Open Korean LLM Leaderboard | 31.2 | Average Score (%) | 26.3 |
| Open Ko-LLM Leaderboard | 31.2 | Average (%) | 26 |
| Open Korean LLM - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
| Open Korean LLM - Ko-IFEval | 9.55 | Strict accuracy, prompt/instruction mean (%) | 7.8 |
| Open Korean LLM - Ko-GPQA-Diamond | 19.19 | Normalized accuracy (%) | 6.4 |
Interactive version: theaggregate.ai/model?slug=custom-kollm-13b-v4 · How It Works · Data refreshed daily, snapshot 2026-09-19.