airoboros-7B-gpt4-1.3: benchmark results
Provider: Jon Durbin. Access: Open.
Aggregate Rating 1264 ± 15, rank #9125 of 10916 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - Medical Genetics | 55 | Accuracy (%) | 62 |
| MMLU-by-task - Anatomy | 47.41 | Accuracy (%) | 61.3 |
| MMLU-by-task - Global Facts | 34 | Accuracy (%) | 58.4 |
| MMLU-by-task - Prehistory | 52.47 | Accuracy (%) | 53.4 |
| MMLU-by-task - HellaSwag | 59.21 | Accuracy (%) | 53.3 |
| MMLU-by-task - Logical Fallacies | 50.92 | Accuracy (%) | 51.4 |
| MMLU-by-task - Public Relations | 53.64 | Accuracy (%) | 51 |
| MMLU-by-task - Jurisprudence | 52.78 | Accuracy (%) | 50 |
| MMLU-by-task - Professional Medicine | 45.59 | Accuracy (%) | 49.2 |
| MMLU-by-task - International Law | 60.33 | Accuracy (%) | 47.6 |
| MMLU-by-task - High School Geography | 52.02 | Accuracy (%) | 47.2 |
| MMLU-by-task - High School World History | 59.92 | Accuracy (%) | 47.2 |
Interactive version: theaggregate.ai/model?slug=airoboros-7b-gpt4-1-3 · How It Works · Data refreshed daily, snapshot 2026-11-08.