EXAONE-4.0.1-32B: benchmark results
Provider: LG AI. Access: Open.
Aggregate Rating 1614 ± 29, rank #893 of 10916 rated models, from 34 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Horangi 4 - Ko-AIME 2025 | 95.65 | Accuracy (%) | 80.8 |
| Horangi 4 - Ko-HalluLens (Nonexistent Entities) | 86 | Refusal rate (%) | 80.3 |
| Horangi 4 - Korean Hate Speech | 64 | Accuracy (%) | 73.1 |
| Horangi 4 - GLP - Mathematical Reasoning | 92.72 | Score (%) | 72.1 |
| Horangi 4 - BFCL | 67.83 | Accuracy (%) | 64.4 |
| Horangi 4 - BigCodeBench (100-task subset) | 51 | Pass rate (%) | 58.7 |
| Horangi 4 - Ko-MT-Bench | 86.75 | Judge rating (1-10, x10) | 55.8 |
| Horangi 4 - HumanEval (100-problem subset) | 95 | Pass rate (%) | 55.3 |
| Horangi 4 - Ko-Moral | 77 | Accuracy (%) | 53.4 |
| Horangi 4 - HRM8K | 89.8 | Accuracy (%) | 49.5 |
| Horangi 4 - Final Score | 65.92 | Score (%) | 47.1 |
| Horangi 4 - KMMLU | 79 | Accuracy (%) | 47.1 |
Interactive version: theaggregate.ai/model?slug=exaone-4-0-1-32b · How It Works · Data refreshed daily, snapshot 2026-11-08.