trurl-2-13B-academic: benchmark results
VoiceLab's Trurl 2 13B Academic, a Polish-and-English fine-tune of Llama 2 13B trained on 855,000 conversational samples (1.7 billion tokens) with a 4,096-token context, excluding MMLU data. Provider: VoiceLab. Released 2023-09-18. Access: Open.
Aggregate Rating 1336 ± 11, rank #1391 of 1635 rated models, from 98 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - Abstract Algebra | 39 | Accuracy (%) | 98 |
| MMLU-by-task - High School Mathematics | 32.22 | Accuracy (%) | 89.1 |
| MMLU-by-task - Sociology | 78.11 | Accuracy (%) | 86.8 |
| MMLU-by-task - College Chemistry | 44 | Accuracy (%) | 86.5 |
| MMLU-by-task - High School Government And Politics | 83.42 | Accuracy (%) | 86.4 |
| MMLU-by-task - High School Physics | 36.42 | Accuracy (%) | 86.3 |
| MMLU-by-task - Anatomy | 51.85 | Accuracy (%) | 85.2 |
| MMLU-by-task - Human Sexuality | 66.41 | Accuracy (%) | 84.7 |
| MMLU-by-task - Elementary Mathematics | 35.45 | Accuracy (%) | 84.2 |
| MMLU-by-task - Virology | 48.19 | Accuracy (%) | 83.9 |
| MMLU-by-task - College Computer Science | 48 | Accuracy (%) | 83.8 |
| MMLU-by-task - Medical Genetics | 62 | Accuracy (%) | 82.4 |
Interactive version: theaggregate.ai/model?slug=trurl-2-13b-academic · How It Works · Data refreshed daily, snapshot 2026-11-05.