SOLAR-10.7B-Instruct-T3Q-ko-solar-dpo-slerp: benchmark results
Provider: Upstage. Access: Open.
Unified ELO 1460 ± 18, rank #1447 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-Helpful | 59.72 | Normalized accuracy (%) | 97.5 |
| Open Korean LLM - KorNAT-Harmless | 73.48 | Normalized accuracy (%) | 96.6 |
| Open Korean LLM - Ko-Winogrande | 68.59 | Accuracy (%) | 93.5 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 47.76 | Exact match, flexible extract (%) | 81.6 |
| Open Korean LLM - KorNAT-CKA | 28.38 | Normalized accuracy (%) | 75.2 |
| Open Korean LLM - Ko-IFEval | 35.55 | Strict accuracy, prompt/instruction mean (%) | 73.2 |
| Open Korean LLM - Ko-GPQA-Diamond | 24.75 | Normalized accuracy (%) | 69.5 |
| Open Korean LLM Leaderboard | 37.46 | Average Score (%) | 62.5 |
| Open Ko-LLM Leaderboard | 37.46 | Average (%) | 62.4 |
| Open Korean LLM - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
Interactive version: theaggregate.ai/model?slug=solar-10-7b-instruct-t3q-ko-solar-dpo-slerp · How It Works · Data refreshed daily, snapshot 2026-09-19.