Llama 2 7B Guanaco Instruct Sharded: benchmark results
Provider: Meta. Access: Open.
Aggregate Rating 1295 ± 15, rank #8813 of 10916 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - Econometrics | 35.09 | Accuracy (%) | 84.6 |
| MMLU-by-task - College Biology | 54.17 | Accuracy (%) | 64.2 |
| MMLU-by-task - Prehistory | 57.1 | Accuracy (%) | 64.1 |
| MMLU-by-task - Management | 67.96 | Accuracy (%) | 62.4 |
| MMLU-by-task - College Mathematics | 33 | Accuracy (%) | 61.7 |
| MMLU-by-task - Virology | 42.77 | Accuracy (%) | 60.3 |
| MMLU-by-task - Philosophy | 57.88 | Accuracy (%) | 59.7 |
| MMLU-by-task - High School US History | 66.67 | Accuracy (%) | 59.3 |
| MMLU-by-task - Clinical Knowledge | 52.45 | Accuracy (%) | 59 |
| MMLU-by-task - Astronomy | 48.03 | Accuracy (%) | 58.3 |
| MMLU-by-task - Marketing | 74.36 | Accuracy (%) | 58.2 |
| MMLU-by-task - World Religions | 70.18 | Accuracy (%) | 57.4 |
Interactive version: theaggregate.ai/model?slug=llama-2-7b-guanaco-instruct-sharded · How It Works · Data refreshed daily, snapshot 2026-11-08.