Mocha-SR-7B-ex: benchmark results
Azazelle's rescaled_sample 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. The card heading retains another label; the exact repository and its recipe define this publication, without a global name alias. Provider: Azazelle. Access: Open.
Aggregate Rating 1389 ± 33, rank #6574 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - GSM8K | 52.39 | Accuracy (%) (5-shot) | 64.2 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 54.72 | MC2 (%) (0-shot) | 63.3 |
| Open LLM Leaderboard v1 - MMLU | 60.87 | Accuracy (%) (5-shot) | 50 |
| Open LLM Leaderboard v1 - ARC Challenge | 59.81 | Normalized accuracy (%) (25-shot) | 45 |
| Open LLM Leaderboard v1 - HellaSwag | 81.24 | Normalized accuracy (%) (10-shot) | 44.1 |
| Open LLM Leaderboard v1 - WinoGrande | 73.09 | Accuracy (%) (5-shot) | 31 |
Interactive version: theaggregate.ai/model?slug=mocha-sr-7b-ex · How It Works · Data refreshed daily, snapshot 2026-11-08.