scarlett-7B: benchmark results
A conversation fine-tune by ajibawa-2023 of Llama-1 on over 10,000 conversation sets covering philosophy, advice and jokes. The card cites Samantha as inspiration; that does not establish checkpoint equivalence. Provider: ajibawa-2023. Access: Open.
Aggregate Rating 1222 ± 16, rank #9465 of 10916 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - HellaSwag | 62.57 | Accuracy (%) | 78.2 |
| MMLU-by-task - TruthfulQA MC1 | 33.29 | Accuracy (%) | 72.3 |
| MMLU-by-task - TruthfulQA MC2 | 48.52 | Accuracy (%) | 67.9 |
| MMLU-by-task - ARC Challenge | 53.67 | Accuracy (%) | 63.9 |
| MMLU-by-task - Global Facts | 32 | Accuracy (%) | 44.3 |
| MMLU-by-task - College Chemistry | 31 | Accuracy (%) | 43.8 |
| MMLU-by-task - High School Physics | 29.14 | Accuracy (%) | 42.7 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 48.52 | MC2 (%) (0-shot) | 41.5 |
| MMLU-by-task - Anatomy | 40.74 | Accuracy (%) | 41.1 |
| MMLU-by-task - High School Government And Politics | 53.37 | Accuracy (%) | 41.1 |
| MMLU-by-task - Professional Medicine | 44.12 | Accuracy (%) | 41 |
| MMLU-by-task - High School European History | 46.06 | Accuracy (%) | 40.9 |
Interactive version: theaggregate.ai/model?slug=scarlett-7b · How It Works · Data refreshed daily, snapshot 2026-11-08.