Arctic Instruct: benchmark results
Provider: Other.
Aggregate Rating 1426 ± 23, rank #4871 of 10916 rated models, from 86 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| HELM MMLU - Computer Security | 84 | EM | 78.2 |
| HELM Lite - LegalBench - Corporate Lobbying | 78.16 | Exact Match (%) | 61.1 |
| HELM NaturalQuestions (Closed) | 38.98 | F1 (%) | 56.7 |
| HELM MMLU - Human Sexuality | 84.73 | EM | 54.5 |
| HELM Lite - GSM8K | 76.8 | Exact Match (%) | 50.6 |
| HELM Lite - MATH Level 1 - Algebra | 78.52 | Equivalent (CoT) (%) | 45.6 |
| HELM MMLU - Sociology | 89.05 | EM | 44.9 |
| HELM Lite - MATH Level 1 - Prealgebra | 72.09 | Equivalent (CoT) (%) | 43.3 |
| HELM Lite - MATH Level 1 - Counting and Probability | 56.41 | Equivalent (CoT) (%) | 42.2 |
| HELM MMLU - World Religions | 85.38 | EM | 41.7 |
| HELM Lite - LegalBench | 58.8 | Exact Match (%) | 40 |
| HELM Lite - MATH Level 1 - Number Theory | 46.67 | Equivalent (CoT) (%) | 40 |
Interactive version: theaggregate.ai/model?slug=arctic-instruct · How It Works · Data refreshed daily, snapshot 2026-11-08.