Noromaid-7B-v0.1.1: benchmark results
A Mistral-7B-v0.1 full fine-tune credited to Undi and IkariDev of NeverSleep. The card describes 1.5 epochs and 1,500 steps, but still labels the model v0.1 inside the v0.1.1 repository; that stale wording does not establish identity with v0.1. Provider: NeverSleep. Access: Open.
Aggregate Rating 1403 ± 33, rank #5913 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - HellaSwag | 84.28 | Normalized accuracy (%) (10-shot) | 65.1 |
| Open LLM Leaderboard v1 - MMLU | 63.44 | Accuracy (%) (5-shot) | 60.7 |
| Open LLM Leaderboard v1 - WinoGrande | 77.9 | Accuracy (%) (5-shot) | 59.3 |
| Open LLM Leaderboard v1 - ARC Challenge | 62.2 | Normalized accuracy (%) (25-shot) | 56.6 |
| Open LLM Leaderboard v1 - GSM8K | 36.85 | Accuracy (%) (5-shot) | 52.2 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 44.3 | MC2 (%) (0-shot) | 26.6 |
Interactive version: theaggregate.ai/model?slug=noromaid-7b-v0-1-1 · How It Works · Data refreshed daily, snapshot 2026-11-08.