stable-vicuna-13B-HF: benchmark results
TheBloke's Hugging Face publication of CarperAI's StableVicuna 13B, whose original release is CarperAI/stable-vicuna-13b-delta. Conversion publication is credited separately from the original model creator. Provider: TheBloke. Access: Open.
Aggregate Rating 1306 ± 15, rank #8627 of 10916 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - College Chemistry | 41 | Accuracy (%) | 77.2 |
| MMLU-by-task - TruthfulQA MC1 | 34.27 | Accuracy (%) | 76.7 |
| MMLU-by-task - Astronomy | 55.92 | Accuracy (%) | 76.5 |
| MMLU-by-task - Human Sexuality | 63.36 | Accuracy (%) | 76.4 |
| MMLU-by-task - High School Mathematics | 30 | Accuracy (%) | 76.1 |
| MMLU-by-task - Abstract Algebra | 33 | Accuracy (%) | 75.6 |
| MMLU-by-task - Machine Learning | 38.39 | Accuracy (%) | 75.5 |
| MMLU-by-task - Virology | 45.78 | Accuracy (%) | 73.2 |
| MMLU-by-task - Formal Logic | 35.71 | Accuracy (%) | 72.7 |
| MMLU-by-task - Security Studies | 61.63 | Accuracy (%) | 71.4 |
| MMLU-by-task - Logical Fallacies | 65.03 | Accuracy (%) | 69.8 |
| MMLU-by-task - World Religions | 74.27 | Accuracy (%) | 68.4 |
Interactive version: theaggregate.ai/model?slug=stable-vicuna-13b-hf · How It Works · Data refreshed daily, snapshot 2026-11-08.