Noromaid-13B-v0.3: benchmark results
A retrained Noromaid role-play release by NeverSleep, credited to Undi and IkariDev, using new datasets and standard Alpaca prompting instead of the earlier custom format. Provider: NeverSleep. Access: Open.
Aggregate Rating 1372 ± 34, rank #7096 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - HellaSwag | 84.42 | Normalized accuracy (%) (10-shot) | 66 |
| Open LLM Leaderboard v1 - ARC Challenge | 62.8 | Normalized accuracy (%) (25-shot) | 59.8 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 50.73 | MC2 (%) (0-shot) | 49.4 |
| Open LLM Leaderboard v1 - GSM8K | 23.05 | Accuracy (%) (5-shot) | 42.7 |
| Open LLM Leaderboard v1 - MMLU | 56.86 | Accuracy (%) (5-shot) | 40.1 |
| Open LLM Leaderboard v1 - WinoGrande | 74.74 | Accuracy (%) (5-shot) | 38.5 |
Interactive version: theaggregate.ai/model?slug=noromaid-13b-v0-3 · How It Works · Data refreshed daily, snapshot 2026-11-08.