Llama 2 70B FP16: benchmark results
TheBloke's Hugging Face publication of Meta's Llama 2 70B, whose original release is meta-llama/Llama-2-70b-hf. Conversion publication is credited separately from the original model creator. Provider: TheBloke. Released 2023-07-18. Access: Open.
Aggregate Rating 1440 ± 17, rank #4188 of 10916 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - High School Macroeconomics | 74.1 | Accuracy (%) | 99.9 |
| MMLU-by-task - High School Statistics | 63.43 | Accuracy (%) | 99.8 |
| MMLU-by-task - Philosophy | 78.78 | Accuracy (%) | 99.7 |
| MMLU-by-task - Sociology | 90.05 | Accuracy (%) | 99.7 |
| MMLU-by-task - Human Sexuality | 87.79 | Accuracy (%) | 99.6 |
| MMLU-by-task - Nutrition | 78.1 | Accuracy (%) | 99.6 |
| MMLU-by-task - College Biology | 84.72 | Accuracy (%) | 99.5 |
| MMLU-by-task - Electrical Engineering | 65.52 | Accuracy (%) | 99.4 |
| MMLU-by-task - Professional Medicine | 75 | Accuracy (%) | 99.2 |
| MMLU-by-task - High School Computer Science | 79 | Accuracy (%) | 99.1 |
| MMLU-by-task - College Computer Science | 60 | Accuracy (%) | 99 |
| MMLU-by-task - Medical Genetics | 74 | Accuracy (%) | 98.9 |
Interactive version: theaggregate.ai/model?slug=llama-2-70b-fp16 · How It Works · Data refreshed daily, snapshot 2026-11-08.