gemma-ko-7B-instruct-v0.71: benchmark results
Provider: Google. Access: API.
Aggregate Rating 1378 ± 27, rank #6900 of 10916 rated models, from 15 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-GPQA-Diamond | 25.76 | Normalized accuracy (%) | 81.3 |
| Open Korean LLM - Ko-EQ-Bench | 24.39 | EQ-Bench score | 69.8 |
| Open Korean LLM - Ko-IFEval | 33.86 | Strict accuracy, prompt/instruction mean (%) | 69.1 |
| Open Korean LLM - KorNAT-Harmless | 63.53 | Normalized accuracy (%) | 63.8 |
| Open LLM Leaderboard v1 - GSM8K | 41.7 | Accuracy (%) (5-shot) | 56.6 |
| Open Korean LLM - KorNAT-CKA | 25.25 | Normalized accuracy (%) | 56.3 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 23.2 | Exact match, flexible extract (%) | 56.1 |
| Open Korean LLM Leaderboard | 35.76 | Average Score (%) | 53.4 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 51.72 | MC2 (%) (0-shot) | 53.1 |
| Open LLM Leaderboard v1 - MMLU | 59.72 | Accuracy (%) (5-shot) | 45.7 |
| Open LLM Leaderboard v1 - ARC Challenge | 55.03 | Normalized accuracy (%) (25-shot) | 32.5 |
| Open LLM Leaderboard v1 - HellaSwag | 77.46 | Normalized accuracy (%) (10-shot) | 30.9 |
Interactive version: theaggregate.ai/model?slug=gemma-ko-7b-instruct-v0-71 · How It Works · Data refreshed daily, snapshot 2026-11-08.