AlphaMonarch-7B: benchmark results
mlabonne's AlphaMonarch 7B, a DPO fine-tune of his NeuralMonarch-7B merge using the OpenHermes 2.5 DPO binarized preference dataset. Provider: mlabonne. Released 2024-03-01. Access: Open.
Aggregate Rating 1447 ± 14, rank #1014 of 1656 rated models, from 53 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-Harmless | 79.63 | Normalized accuracy (%) | 98.7 |
| Open LLM Leaderboard v1 - HellaSwag | 89.18 | Normalized accuracy (%) (10-shot) | 98.3 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 77.91 | MC2 (%) (0-shot) | 97.6 |
| Open LLM Leaderboard v1 - ARC Challenge | 73.04 | Normalized accuracy (%) (25-shot) | 95.2 |
| Open LLM Leaderboard v1 - WinoGrande | 84.69 | Accuracy (%) (5-shot) | 94.4 |
| Open Korean LLM - KorNAT-Helpful | 52.23 | Normalized accuracy (%) | 92.3 |
| YALL Nous Leaderboard | 57.53 | Average | 92.1 |
| Open Chinese LLM - TruthfulQA MC | 62.69 | Accuracy (%) | 92 |
| Open Korean LLM - KorNAT-CKA | 33.16 | Normalized accuracy (%) | 90.9 |
| Open Korean LLM Leaderboard | 42.34 | Average Score (%) | 86.1 |
| Open Ko-LLM Leaderboard | 42.34 | Average (%) | 85.9 |
| Open LLM Leaderboard v1 - GSM8K | 66.72 | Accuracy (%) (5-shot) | 82.3 |
Interactive version: theaggregate.ai/model?slug=alphamonarch-7b · How It Works · Data refreshed daily, snapshot 2026-10-20.