Cerebras-GPT-13B: benchmark results
Provider: Cerebras. Released 2023-03-28.
Aggregate Rating 1115 ± 15, rank #10578 of 10916 rated models, from 81 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - High School Statistics | 45.37 | Accuracy (%) | 75.3 |
| MMLU-by-task - College Mathematics | 34 | Accuracy (%) | 69.2 |
| MMLU-by-task - Moral Scenarios | 27.26 | Accuracy (%) | 59.8 |
| MMLU-by-task - Formal Logic | 30.16 | Accuracy (%) | 48.8 |
| MMLU-by-task - Security Studies | 35.51 | Accuracy (%) | 32.8 |
| MMLU-by-task - College Biology | 29.17 | Accuracy (%) | 32 |
| MMLU-by-task - High School Macroeconomics | 31.79 | Accuracy (%) | 29 |
| MMLU-by-task - High School Microeconomics | 27.73 | Accuracy (%) | 27.9 |
| MMLU-by-task - Virology | 30.72 | Accuracy (%) | 26.8 |
| MMLU-by-task - Machine Learning | 27.68 | Accuracy (%) | 26.7 |
| MMLU-by-task - High School World History | 27.85 | Accuracy (%) | 26.2 |
| MMLU-by-task - Miscellaneous | 28.35 | Accuracy (%) | 25.1 |
Interactive version: theaggregate.ai/model?slug=cerebras-gpt-13b · How It Works · Data refreshed daily, snapshot 2026-11-08.