Moko-DARE: benchmark results
Azazelle's DARE-TIES merge, whose published recipe uses mistralai/Mistral-7B-v0.1 as its base and Open-Orca/Mistral-7B-OpenOrca, akjindal53244/Mistral-7B-v0.1-Open-Platypus, WizardLM/WizardMath-7B-V1.1 as merge inputs. Its published Mistral configuration specifies 32 decoder layers. Provider: Azazelle. Access: Open.
Aggregate Rating 1384 ± 35, rank #6731 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 52.17 | MC2 (%) (0-shot) | 54.8 |
| Open LLM Leaderboard v1 - MMLU | 61.94 | Accuracy (%) (5-shot) | 53.4 |
| Open LLM Leaderboard v1 - HellaSwag | 82.08 | Normalized accuracy (%) (10-shot) | 48.7 |
| Open LLM Leaderboard v1 - ARC Challenge | 60.58 | Normalized accuracy (%) (25-shot) | 48.4 |
| Open LLM Leaderboard v1 - WinoGrande | 75.14 | Accuracy (%) (5-shot) | 40.5 |
| Open LLM Leaderboard v1 - GSM8K | 5.08 | Accuracy (%) (5-shot) | 23.7 |
Interactive version: theaggregate.ai/model?slug=moko-dare · How It Works · Data refreshed daily, snapshot 2026-11-08.