AISquare-Instruct-yi-ko-6B-v0.9.31: benchmark results
Provider: Other. Access: Open.
Unified ELO 1392 ± 17, rank #1970 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-Harmless | 66.27 | Normalized accuracy (%) | 85.1 |
| Open Korean LLM - KorNAT-CKA | 29.61 | Normalized accuracy (%) | 80.6 |
| Open Korean LLM - Ko-Winogrande | 60.3 | Accuracy (%) | 53 |
| Open Korean LLM Leaderboard | 33.1 | Average Score (%) | 38.3 |
| Open Ko-LLM Leaderboard | 33.1 | Average (%) | 38.1 |
| Open Korean LLM - Ko-IFEval | 23.64 | Strict accuracy, prompt/instruction mean (%) | 27.7 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 6.29 | Exact match, flexible extract (%) | 27.4 |
| Open Korean LLM - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
| Open Korean LLM - KorNAT-Helpful | 43.5 | Normalized accuracy (%) | 23.8 |
| Open Korean LLM - Ko-GPQA-Diamond | 19.7 | Normalized accuracy (%) | 9.6 |
Interactive version: theaggregate.ai/model?slug=aisquare-instruct-yi-ko-6b-v0-9-31 · How It Works · Data refreshed daily, snapshot 2026-09-19.