test_instruction: benchmark results
Provider: Other. Access: Open.
Unified ELO 1424 ± 17, rank #1720 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-EQ-Bench | 17.78 | EQ-Bench score | 63.8 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 23.65 | Exact match, flexible extract (%) | 56.7 |
| Open Korean LLM - KorNAT-Helpful | 47.25 | Normalized accuracy (%) | 54.7 |
| Open Ko-LLM Leaderboard | 35.74 | Average (%) | 53.3 |
| Open Korean LLM Leaderboard | 35.74 | Average Score (%) | 53.3 |
| Open Korean LLM - Ko-IFEval | 28.37 | Strict accuracy, prompt/instruction mean (%) | 47.5 |
| Open Korean LLM - KorNAT-Harmless | 62.29 | Normalized accuracy (%) | 41.3 |
| Open Korean LLM - Ko-Winogrande | 57.14 | Accuracy (%) | 29.8 |
| Open Korean LLM - KorNAT-CKA | 17.66 | Normalized accuracy (%) | 15.7 |
| Open Korean LLM - Ko-GPQA-Diamond | 19.7 | Normalized accuracy (%) | 9.6 |
Interactive version: theaggregate.ai/model?slug=test-instruction · How It Works · Data refreshed daily, snapshot 2026-09-19.