Mini_DPO_test02: benchmark results
Provider: Other. Access: Open.
Aggregate Rating 1388 ± 29, rank #6603 of 10916 rated models, from 15 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-Helpful | 52.55 | Normalized accuracy (%) | 93.4 |
| Open Korean LLM - KorNAT-Harmless | 68.57 | Normalized accuracy (%) | 92.3 |
| Open Korean LLM - KorNAT-CKA | 26.66 | Normalized accuracy (%) | 66 |
| Open LLM Leaderboard v1 - WinoGrande | 78.37 | Accuracy (%) (5-shot) | 62.7 |
| Open LLM Leaderboard v1 - HellaSwag | 83.89 | Normalized accuracy (%) (10-shot) | 62.3 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 24.56 | Exact match, flexible extract (%) | 57.7 |
| Open LLM Leaderboard v1 - MMLU | 61.9 | Accuracy (%) (5-shot) | 53.2 |
| Open Korean LLM - Ko-Winogrande | 60.22 | Accuracy (%) | 52.3 |
| Open LLM Leaderboard v1 - GSM8K | 35.03 | Accuracy (%) (5-shot) | 50.8 |
| Open LLM Leaderboard v1 - ARC Challenge | 59.73 | Normalized accuracy (%) (25-shot) | 44.7 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 48.47 | MC2 (%) (0-shot) | 41.3 |
| Open Korean LLM Leaderboard | 33.29 | Average Score (%) | 40.6 |
Interactive version: theaggregate.ai/model?slug=mini-dpo-test02 · How It Works · Data refreshed daily, snapshot 2026-11-08.