ko-llama2-finetune-ex4: benchmark results
Provider: Other. Access: Open.
Unified ELO 1297 ± 17, rank #2467 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-Harmless | 62.41 | Normalized accuracy (%) | 43.7 |
| Open Korean LLM - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
| Open Korean LLM - Ko-GPQA-Diamond | 21.21 | Normalized accuracy (%) | 24 |
| Open Korean LLM - KorNAT-Helpful | 43.15 | Normalized accuracy (%) | 21.6 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 2.65 | Exact match, flexible extract (%) | 19.7 |
| Open Korean LLM - Ko-Winogrande | 52.57 | Accuracy (%) | 10 |
| Open Korean LLM - KorNAT-CKA | 15.33 | Normalized accuracy (%) | 9.1 |
| Open Korean LLM - Ko-IFEval | 11.6 | Strict accuracy, prompt/instruction mean (%) | 8.6 |
| Open Korean LLM Leaderboard | 28.19 | Average Score (%) | 5.4 |
| Open Ko-LLM Leaderboard | 28.19 | Average (%) | 5.3 |
Interactive version: theaggregate.ai/model?slug=ko-llama2-finetune-ex4 · How It Works · Data refreshed daily, snapshot 2026-09-19.