SJ-SOLAR-10.7B-DPO: benchmark results
Provider: Other. Access: Open.
Aggregate Rating 1448 ± 30, rank #3844 of 10916 rated models, from 15 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-Harmless | 71.94 | Normalized accuracy (%) | 95.1 |
| Open LLM Leaderboard v1 - WinoGrande | 84.21 | Accuracy (%) (5-shot) | 92.4 |
| Open Korean LLM - Ko-Winogrande | 67.32 | Accuracy (%) | 90.8 |
| Open LLM Leaderboard v1 - MMLU | 66.73 | Accuracy (%) (5-shot) | 88.1 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 67.74 | MC2 (%) (0-shot) | 86.5 |
| Open LLM Leaderboard v1 - HellaSwag | 86.95 | Normalized accuracy (%) (10-shot) | 83.7 |
| Open Korean LLM - KorNAT-CKA | 29.61 | Normalized accuracy (%) | 80.6 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 45.56 | Exact match, flexible extract (%) | 80.4 |
| Open LLM Leaderboard v1 - ARC Challenge | 68.26 | Normalized accuracy (%) (25-shot) | 79.8 |
| Open LLM Leaderboard v1 - GSM8K | 62.09 | Accuracy (%) (5-shot) | 75.1 |
| Open Korean LLM Leaderboard | 39.24 | Average Score (%) | 70.2 |
| Open Korean LLM - KorNAT-Helpful | 48.38 | Normalized accuracy (%) | 67.5 |
Interactive version: theaggregate.ai/model?slug=sj-solar-10-7b-dpo · How It Works · Data refreshed daily, snapshot 2026-11-08.