TinyKo-v5-a: benchmark results
Provider: Other. Access: Open.
Unified ELO 1243 ± 18, rank #2582 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-Harmless | 64.51 | Normalized accuracy (%) | 74.7 |
| Open Korean LLM - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 1.59 | Exact match, flexible extract (%) | 14.5 |
| Open Korean LLM - Ko-IFEval | 15.93 | Strict accuracy, prompt/instruction mean (%) | 10.8 |
| Open Korean LLM - KorNAT-Helpful | 40.52 | Normalized accuracy (%) | 7.8 |
| Open Korean LLM - Ko-GPQA-Diamond | 18.69 | Normalized accuracy (%) | 4.1 |
| Open Korean LLM - KorNAT-CKA | 10.12 | Normalized accuracy (%) | 1.8 |
| Open Ko-LLM Leaderboard | 25.8 | Average (%) | 0.8 |
| Open Korean LLM Leaderboard | 25.8 | Average Score (%) | 0.8 |
| Open Korean LLM - Ko-Winogrande | 48.54 | Accuracy (%) | 0.5 |
Interactive version: theaggregate.ai/model?slug=tinyko-v5-a · How It Works · Data refreshed daily, snapshot 2026-09-19.