Llama 2 22B: benchmark results
Provider: chargoddard. Access: Open.
Aggregate Rating 1338 ± 15, rank #7998 of 10916 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - High School Chemistry | 46.8 | Accuracy (%) | 89.7 |
| MMLU-by-task - Computer Security | 72 | Accuracy (%) | 88.7 |
| MMLU-by-task - Nutrition | 63.73 | Accuracy (%) | 87 |
| MMLU-by-task - Anatomy | 51.85 | Accuracy (%) | 85.2 |
| MMLU-by-task - Jurisprudence | 73.15 | Accuracy (%) | 84 |
| MMLU-by-task - College Computer Science | 48 | Accuracy (%) | 83.8 |
| MMLU-by-task - High School Mathematics | 30.74 | Accuracy (%) | 81.5 |
| MMLU-by-task - Management | 73.79 | Accuracy (%) | 79.4 |
| MMLU-by-task - High School Computer Science | 58 | Accuracy (%) | 78.9 |
| MMLU-by-task - Moral Disputes | 63.58 | Accuracy (%) | 78.9 |
| MMLU-by-task - College Biology | 59.72 | Accuracy (%) | 77.1 |
| MMLU-by-task - High School US History | 75.98 | Accuracy (%) | 76.9 |
Interactive version: theaggregate.ai/model?slug=llama-2-22b · How It Works · Data refreshed daily, snapshot 2026-11-08.