Tippy-Toppy-7B: benchmark results
Azazelle's DARE-TIES merge, whose published recipe uses mistralai/Mistral-7B-v0.1 as its base and Undi95/Toppy-M-7B, PistachioAlt/Noromaid-Bagel-7B-Slerp, OpenPipe/mistral-ft-optimized-1227 as merge inputs. Its published Mistral configuration specifies 32 decoder layers. Provider: Azazelle. Access: Open.
Aggregate Rating 1434 ± 33, rank #4446 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - MMLU | 65.49 | Accuracy (%) (5-shot) | 84.6 |
| Open LLM Leaderboard v1 - GSM8K | 64.67 | Accuracy (%) (5-shot) | 78.7 |
| Open LLM Leaderboard v1 - HellaSwag | 85.88 | Normalized accuracy (%) (10-shot) | 77.3 |
| Open LLM Leaderboard v1 - ARC Challenge | 66.89 | Normalized accuracy (%) (25-shot) | 75.1 |
| Open LLM Leaderboard v1 - WinoGrande | 78.85 | Accuracy (%) (5-shot) | 65.9 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 55.7 | MC2 (%) (0-shot) | 65.8 |
Interactive version: theaggregate.ai/model?slug=tippy-toppy-7b · How It Works · Data refreshed daily, snapshot 2026-11-08.