SOLAR-10.7B-Instruct-T3Q-ko-solar-dpo-linear: benchmark results
Provider: Upstage. Access: Open.
Unified ELO 1479 ± 18, rank #1298 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-Helpful | 55.27 | Normalized accuracy (%) | 95.8 |
| Open Korean LLM - KorNAT-Harmless | 71.89 | Normalized accuracy (%) | 94.9 |
| Open Korean LLM - Ko-Winogrande | 69.14 | Accuracy (%) | 94.5 |
| Open Korean LLM - KorNAT-CKA | 31.74 | Normalized accuracy (%) | 87.4 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 52.69 | Exact match, flexible extract (%) | 86.1 |
| Open Korean LLM - Ko-IFEval | 36.2 | Strict accuracy, prompt/instruction mean (%) | 75.1 |
| Open Ko-LLM Leaderboard | 38.95 | Average (%) | 69.1 |
| Open Korean LLM Leaderboard | 38.95 | Average Score (%) | 69.1 |
| Open Korean LLM - Ko-EQ-Bench | 20.16 | EQ-Bench score | 66 |
| Open Korean LLM - Ko-GPQA-Diamond | 22.22 | Normalized accuracy (%) | 35 |
Interactive version: theaggregate.ai/model?slug=solar-10-7b-instruct-t3q-ko-solar-dpo-linear · How It Works · Data refreshed daily, snapshot 2026-09-19.