fairseq-dense-13B: benchmark results
Provider: Meta. Access: Open.
Aggregate Rating 1147 ± 16, rank #10013 of 10916 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - High School Physics | 31.13 | Accuracy (%) | 56.1 |
| MMLU-by-task - High School Statistics | 36.11 | Accuracy (%) | 44.3 |
| MMLU-by-task - Professional Accounting | 31.91 | Accuracy (%) | 42.1 |
| MMLU-by-task - HellaSwag | 55.95 | Accuracy (%) | 41.8 |
| MMLU-by-task - Moral Scenarios | 24.69 | Accuracy (%) | 38.9 |
| MMLU-by-task - High School Computer Science | 36 | Accuracy (%) | 38.2 |
| MMLU-by-task - Global Facts | 31 | Accuracy (%) | 36.8 |
| MMLU-by-task - High School Microeconomics | 33.61 | Accuracy (%) | 36.3 |
| MMLU-by-task - Clinical Knowledge | 33.96 | Accuracy (%) | 34.9 |
| MMLU-by-task - Professional Psychology | 28.76 | Accuracy (%) | 32.6 |
| MMLU-by-task - Computer Security | 34 | Accuracy (%) | 31.7 |
| MMLU-by-task - Prehistory | 30.25 | Accuracy (%) | 30.9 |
Interactive version: theaggregate.ai/model?slug=fairseq-dense-13b · How It Works · Data refreshed daily, snapshot 2026-11-08.