alpaca-lora-65B-HF: benchmark results
TheBloke's Hugging Face publication of changsung's alpaca-lora-65B, whose original release is chansung/alpaca-lora-65b. Conversion publication is credited separately from the original model creator. Provider: TheBloke. Access: Open.
Aggregate Rating 1389 ± 15, rank #6545 of 10916 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - College Chemistry | 51 | Accuracy (%) | 98.6 |
| MMLU-by-task - Medical Genetics | 71 | Accuracy (%) | 96.7 |
| MMLU-by-task - Management | 82.52 | Accuracy (%) | 96.4 |
| MMLU-by-task - Machine Learning | 50.89 | Accuracy (%) | 95.5 |
| MMLU-by-task - Computer Security | 76 | Accuracy (%) | 94.7 |
| MMLU-by-task - High School Statistics | 55.09 | Accuracy (%) | 93.4 |
| MMLU-by-task - Human Sexuality | 77.86 | Accuracy (%) | 93.1 |
| MMLU-by-task - Moral Disputes | 73.99 | Accuracy (%) | 93.1 |
| MMLU-by-task - Moral Scenarios | 47.93 | Accuracy (%) | 93.1 |
| MMLU-by-task - HellaSwag | 67.2 | Accuracy (%) | 93 |
| MMLU-by-task - Logical Fallacies | 77.91 | Accuracy (%) | 93 |
| MMLU-by-task - Virology | 52.41 | Accuracy (%) | 92.8 |
Interactive version: theaggregate.ai/model?slug=alpaca-lora-65b-hf · How It Works · Data refreshed daily, snapshot 2026-11-08.