KIT-5.8B: benchmark results
Provider: Other. Access: Open.
Unified ELO 1358 ± 18, rank #2184 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-CKA | 26.65 | Normalized accuracy (%) | 65.9 |
| Open Korean LLM - Ko-GPQA-Diamond | 24.24 | Normalized accuracy (%) | 62.4 |
| Open Korean LLM - Ko-IFEval | 26.33 | Strict accuracy, prompt/instruction mean (%) | 38.6 |
| Open Korean LLM Leaderboard | 31.08 | Average Score (%) | 24.8 |
| Open Ko-LLM Leaderboard | 31.08 | Average (%) | 24.5 |
| Open Korean LLM - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
| Open Korean LLM - Ko-Winogrande | 53.91 | Accuracy (%) | 14.6 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 0.91 | Exact match, flexible extract (%) | 11.5 |
| Open Korean LLM - KorNAT-Harmless | 59.49 | Normalized accuracy (%) | 6.7 |
| Open Korean LLM - KorNAT-Helpful | 38.73 | Normalized accuracy (%) | 4 |
Interactive version: theaggregate.ai/model?slug=kit-5-8b · How It Works · Data refreshed daily, snapshot 2026-09-19.