ko-gemma-7B-sft-dpo-v1.0: benchmark results
Provider: Other. Access: Open.
Unified ELO 1422 ± 17, rank #1737 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-Helpful | 48.8 | Normalized accuracy (%) | 72.8 |
| Open Korean LLM - KorNAT-Harmless | 64.12 | Normalized accuracy (%) | 70.8 |
| Open Korean LLM - KorNAT-CKA | 26.46 | Normalized accuracy (%) | 64.6 |
| Open Korean LLM - Ko-EQ-Bench | 14.25 | EQ-Bench score | 60.1 |
| Open Korean LLM - Ko-IFEval | 30.87 | Strict accuracy, prompt/instruction mean (%) | 58.4 |
| Open Korean LLM Leaderboard | 35.2 | Average Score (%) | 50.3 |
| Open Ko-LLM Leaderboard | 35.2 | Average (%) | 50.1 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 9.25 | Exact match, flexible extract (%) | 32.5 |
| Open Korean LLM - Ko-Winogrande | 52.57 | Accuracy (%) | 10 |
Interactive version: theaggregate.ai/model?slug=ko-gemma-7b-sft-dpo-v1-0 · How It Works · Data refreshed daily, snapshot 2026-09-19.