Llama 2 70B Ensemble V4: benchmark results
Provider: Meta.
Aggregate Rating 1427 ± 21, rank #4807 of 10916 rated models, from 62 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - College Biology | 85.42 | Accuracy (%) | 100 |
| MMLU-by-task - College Mathematics | 45 | Accuracy (%) | 99.8 |
| MMLU-by-task - Professional Law | 59.19 | Accuracy (%) | 99.8 |
| MMLU-by-task - Human Aging | 81.61 | Accuracy (%) | 99.7 |
| MMLU-by-task - Moral Scenarios | 65.03 | Accuracy (%) | 99.7 |
| MMLU-by-task - High School Microeconomics | 79.41 | Accuracy (%) | 99.6 |
| MMLU-by-task - High School Government And Politics | 94.82 | Accuracy (%) | 99.5 |
| MMLU-by-task - Formal Logic | 52.38 | Accuracy (%) | 99.4 |
| MMLU-by-task - High School European History | 86.06 | Accuracy (%) | 99.4 |
| MMLU-by-task - Machine Learning | 58.93 | Accuracy (%) | 99.4 |
| MMLU-by-task - ARC Challenge | 68.09 | Accuracy (%) | 99.1 |
| MMLU-by-task - Elementary Mathematics | 47.09 | Accuracy (%) | 99.1 |
Interactive version: theaggregate.ai/model?slug=llama-2-70b-ensemble-v4 · How It Works · Data refreshed daily, snapshot 2026-11-08.