gemma-ko-7B-it-v0.40: benchmark results
Provider: Google. Access: API.
Unified ELO 1464 ± 17, rank #1418 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-EQ-Bench | 35.06 | EQ-Bench score | 84.9 |
| Open Korean LLM - KorNAT-CKA | 27.51 | Normalized accuracy (%) | 70.4 |
| Open Ko-LLM Leaderboard | 39.15 | Average (%) | 70.3 |
| Open Korean LLM Leaderboard | 39.15 | Average Score (%) | 70.3 |
| Open Korean LLM - Ko-GPQA-Diamond | 24.24 | Normalized accuracy (%) | 62.4 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 25.25 | Exact match, flexible extract (%) | 58.6 |
| Open Korean LLM - Ko-IFEval | 29.33 | Strict accuracy, prompt/instruction mean (%) | 51.3 |
| Open Korean LLM - KorNAT-Harmless | 61.44 | Normalized accuracy (%) | 28.6 |
| Open Korean LLM - KorNAT-Helpful | 43.77 | Normalized accuracy (%) | 25.3 |
| Open Korean LLM - Ko-Winogrande | 55.25 | Accuracy (%) | 19.4 |
Interactive version: theaggregate.ai/model?slug=gemma-ko-7b-it-v0-40 · How It Works · Data refreshed daily, snapshot 2026-09-19.