solar-ko-common-merged: benchmark results
Provider: Upstage. Access: Open.
Unified ELO 1451 ± 17, rank #1521 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-CKA | 29.78 | Normalized accuracy (%) | 81.3 |
| Open Korean LLM - Ko-EQ-Bench | 28.45 | EQ-Bench score | 76.1 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 36.92 | Exact match, flexible extract (%) | 70.5 |
| Open Ko-LLM Leaderboard | 37.45 | Average (%) | 62.2 |
| Open Korean LLM Leaderboard | 37.45 | Average Score (%) | 62.2 |
| Open Korean LLM - Ko-IFEval | 32.38 | Strict accuracy, prompt/instruction mean (%) | 62.1 |
| Open Korean LLM - KorNAT-Helpful | 45.32 | Normalized accuracy (%) | 35.8 |
| Open Korean LLM - KorNAT-Harmless | 61.5 | Normalized accuracy (%) | 30 |
| Open Korean LLM - Ko-GPQA-Diamond | 21.72 | Normalized accuracy (%) | 29.4 |
| Open Korean LLM - Ko-Winogrande | 57.06 | Accuracy (%) | 29.3 |
Interactive version: theaggregate.ai/model?slug=solar-ko-common-merged · How It Works · Data refreshed daily, snapshot 2026-09-19.