juud-Mistral-7B-dpo: benchmark results
Provider: Other. Released 2024-02-07. Access: Open.
Aggregate Rating 1362 ± 15, rank #1361 of 1656 rated models, from 24 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-Harmless | 64.6 | Normalized accuracy (%) | 75.3 |
| Open LLM Leaderboard v1 - ARC Challenge | 66.81 | Normalized accuracy (%) (25-shot) | 74.8 |
| Open LLM Leaderboard v1 - HellaSwag | 84.89 | Normalized accuracy (%) (10-shot) | 69.9 |
| Open Korean LLM - KorNAT-CKA | 27.18 | Normalized accuracy (%) | 68 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 28.13 | Exact match, flexible extract (%) | 62.5 |
| Open LLM Leaderboard v1 - WinoGrande | 78.3 | Accuracy (%) (5-shot) | 62 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 53.51 | MC2 (%) (0-shot) | 59.8 |
| Open Korean LLM - Ko-EQ-Bench | 12.03 | EQ-Bench score | 58 |
| Open LLM Leaderboard v1 - MMLU | 63.03 | Accuracy (%) (5-shot) | 58 |
| Open Korean LLM - Ko-IFEval | 29.74 | Strict accuracy, prompt/instruction mean (%) | 52.6 |
| Open Chinese LLM - TruthfulQA MC | 53.91 | Accuracy (%) | 51.9 |
| Open Korean LLM Leaderboard | 33.29 | Average Score (%) | 40.7 |
Interactive version: theaggregate.ai/model?slug=juud-mistral-7b-dpo · How It Works · Data refreshed daily, snapshot 2026-10-20.