Synthia-70B-v1.2B: benchmark results
Provider: migtissera. Access: Open.
Aggregate Rating 1434 ± 16, rank #4477 of 10916 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - Formal Logic | 51.59 | Accuracy (%) | 99.2 |
| MMLU-by-task - Philosophy | 78.46 | Accuracy (%) | 99.2 |
| MMLU-by-task - Human Aging | 80.72 | Accuracy (%) | 99 |
| MMLU-by-task - Moral Disputes | 79.19 | Accuracy (%) | 98.9 |
| MMLU-by-task - Virology | 55.42 | Accuracy (%) | 98.8 |
| MMLU-by-task - Marketing | 91.03 | Accuracy (%) | 98.7 |
| MMLU-by-task - Anatomy | 64.44 | Accuracy (%) | 98.4 |
| MMLU-by-task - HellaSwag | 68.86 | Accuracy (%) | 98.4 |
| MMLU-by-task - High School Geography | 87.88 | Accuracy (%) | 97.8 |
| MMLU-by-task - Medical Genetics | 72 | Accuracy (%) | 97.8 |
| MMLU-by-task - College Chemistry | 50 | Accuracy (%) | 97.4 |
| MMLU-by-task - Security Studies | 79.59 | Accuracy (%) | 97.4 |
Interactive version: theaggregate.ai/model?slug=synthia-70b-v1-2b · How It Works · Data refreshed daily, snapshot 2026-11-08.