VAETKI: benchmark results
Provider: Other.
Aggregate Rating 1559 ± 27, rank #1569 of 10916 rated models, from 35 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Horangi 4 - Korean Hate Speech | 64 | Accuracy (%) | 73.1 |
| Horangi 4 - KoBBQ | 82 | Accuracy (%) | 41.3 |
| Horangi 4 - Ko-HalluLens (Nonexistent Entities) | 70 | Refusal rate (%) | 40.4 |
| Horangi 4 - IFEval-Ko | 81.25 | Instruction-following accuracy (%) | 38.9 |
| Horangi 4 - KoBALT-700 (Syntax) | 49 | Accuracy (%) | 38 |
| Horangi 4 - ALT Average | 70.12 | Score (%) | 37.5 |
| Horangi 4 - HAE-RAE Bench (without Reading Comprehension) | 80.81 | Accuracy (%) | 32.2 |
| Horangi 4 - KMMLU-Pro | 57 | Accuracy (%) | 29.8 |
| Horangi 4 - Ko-Moral | 70 | Accuracy (%) | 29.3 |
| Horangi 4 - GLP - Specialized Knowledge | 33 | Score (%) | 28.8 |
| Horangi 4 - Ko-HalluLens (WikiQA) | 18 | Correct answer rate (%) | 28.4 |
| Horangi 4 - ALT - Hallucination Prevention | 53.33 | Score (%) | 27.4 |
Interactive version: theaggregate.ai/model?slug=vaetki · How It Works · Data refreshed daily, snapshot 2026-11-08.