ko-llama2-finetune-ex3: benchmark results
Provider: Other. Access: Open.
Unified ELO 1286 ± 18, rank #2493 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-Harmless | 62.54 | Normalized accuracy (%) | 45.5 |
| Open Korean LLM - Ko-GPQA-Diamond | 21.72 | Normalized accuracy (%) | 29.4 |
| Open Korean LLM - KorNAT-Helpful | 43.73 | Normalized accuracy (%) | 25.1 |
| Open Korean LLM - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 4.47 | Exact match, flexible extract (%) | 23 |
| Open Korean LLM - KorNAT-CKA | 15.46 | Normalized accuracy (%) | 9.2 |
| Open Korean LLM - Ko-Winogrande | 52.41 | Accuracy (%) | 9.1 |
| Open Korean LLM - Ko-IFEval | 2.2 | Strict accuracy, prompt/instruction mean (%) | 4.1 |
| Open Korean LLM Leaderboard | 27.88 | Average Score (%) | 3.3 |
| Open Ko-LLM Leaderboard | 27.88 | Average (%) | 3.2 |
Interactive version: theaggregate.ai/model?slug=ko-llama2-finetune-ex3 · How It Works · Data refreshed daily, snapshot 2026-09-19.