UndiMix-v1-13B: benchmark results
Provider: Undi95. Access: Open.
Aggregate Rating 1344 ± 15, rank #7865 of 10916 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - Human Aging | 68.16 | Accuracy (%) | 88.3 |
| MMLU-by-task - Security Studies | 67.35 | Accuracy (%) | 87.5 |
| MMLU-by-task - College Computer Science | 49 | Accuracy (%) | 86.2 |
| MMLU-by-task - Computer Security | 71 | Accuracy (%) | 86.2 |
| MMLU-by-task - Public Relations | 66.36 | Accuracy (%) | 85.5 |
| MMLU-by-task - High School Biology | 68.39 | Accuracy (%) | 85.4 |
| MMLU-by-task - International Law | 75.21 | Accuracy (%) | 84.2 |
| MMLU-by-task - Jurisprudence | 73.15 | Accuracy (%) | 84 |
| MMLU-by-task - Human Sexuality | 65.65 | Accuracy (%) | 83.1 |
| MMLU-by-task - HellaSwag | 63.36 | Accuracy (%) | 82.5 |
| MMLU-by-task - Miscellaneous | 76.76 | Accuracy (%) | 82.2 |
| MMLU-by-task - Abstract Algebra | 34 | Accuracy (%) | 82 |
Interactive version: theaggregate.ai/model?slug=undimix-v1-13b · How It Works · Data refreshed daily, snapshot 2026-11-08.