UndiMix-v4-13B: benchmark results
Provider: Undi95. Access: Open.
Aggregate Rating 1351 ± 15, rank #7663 of 10916 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - Human Aging | 70.85 | Accuracy (%) | 92.1 |
| MMLU-by-task - High School Mathematics | 32.96 | Accuracy (%) | 91.8 |
| MMLU-by-task - Abstract Algebra | 36 | Accuracy (%) | 91.7 |
| MMLU-by-task - Computer Security | 73 | Accuracy (%) | 90.5 |
| MMLU-by-task - Clinical Knowledge | 63.77 | Accuracy (%) | 89.7 |
| MMLU-by-task - US Foreign Policy | 85 | Accuracy (%) | 89.3 |
| MMLU-by-task - Jurisprudence | 75.93 | Accuracy (%) | 88.9 |
| MMLU-by-task - Logical Fallacies | 71.17 | Accuracy (%) | 88 |
| MMLU-by-task - Moral Scenarios | 44.13 | Accuracy (%) | 87.8 |
| MMLU-by-task - HellaSwag | 64.45 | Accuracy (%) | 87.5 |
| MMLU-by-task - College Medicine | 54.91 | Accuracy (%) | 86.5 |
| MMLU-by-task - College Computer Science | 49 | Accuracy (%) | 86.2 |
Interactive version: theaggregate.ai/model?slug=undimix-v4-13b · How It Works · Data refreshed daily, snapshot 2026-11-08.