SOLAR-10.7B-sft-dpo-v1: benchmark results
Provider: Upstage. Access: Open.
Unified ELO 1463 ± 17, rank #1422 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-Winogrande | 75.14 | Accuracy (%) | 99.6 |
| Open Korean LLM - KorNAT-Harmless | 79.84 | Normalized accuracy (%) | 99 |
| Open Korean LLM - KorNAT-Helpful | 68.62 | Normalized accuracy (%) | 98.8 |
| Open Korean LLM - Ko-GPQA-Diamond | 27.78 | Normalized accuracy (%) | 95 |
| Open Korean LLM - KorNAT-CKA | 28.78 | Normalized accuracy (%) | 77.1 |
| Open Ko-LLM Leaderboard | 38.49 | Average (%) | 67.4 |
| Open Korean LLM Leaderboard | 38.49 | Average Score (%) | 67.4 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 20.62 | Exact match, flexible extract (%) | 52.6 |
| Open Korean LLM - Ko-EQ-Bench | 3.7 | EQ-Bench score | 49.8 |
| Open Korean LLM - Ko-IFEval | 26.77 | Strict accuracy, prompt/instruction mean (%) | 40.3 |
Interactive version: theaggregate.ai/model?slug=solar-10-7b-sft-dpo-v1 · How It Works · Data refreshed daily, snapshot 2026-09-19.