gemma-2-9b_dpo-v4-hq: benchmark results
Provider: Google. Released 2024-08-14. Access: API.
Unified ELO 1598 ± 17, rank #636 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-EQ-Bench | 60.19 | EQ-Bench score | 99.1 |
| Open Korean LLM - Ko-IFEval | 67.89 | Strict accuracy, prompt/instruction mean (%) | 99 |
| Open Ko-LLM Leaderboard | 51.82 | Average (%) | 98.4 |
| Open Korean LLM Leaderboard | 51.82 | Average Score (%) | 98.4 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 69.6 | Exact match, flexible extract (%) | 98.2 |
| Open Korean LLM - KorNAT-CKA | 33.72 | Normalized accuracy (%) | 93 |
| Open Korean LLM - KorNAT-Helpful | 51 | Normalized accuracy (%) | 88.5 |
| Open Korean LLM - KorNAT-Harmless | 67 | Normalized accuracy (%) | 87.6 |
| Open Korean LLM - Ko-Winogrande | 64.64 | Accuracy (%) | 79.8 |
| Open Korean LLM - Ko-GPQA-Diamond | 23.74 | Normalized accuracy (%) | 55.6 |
Interactive version: theaggregate.ai/model?slug=gemma-2-9b-dpo-v4-hq · How It Works · Data refreshed daily, snapshot 2026-09-19.