EXAONE 4.0 32B: benchmark results
Provider: LG AI. Released 2025-07-15. Access: Open.
Aggregate Rating 1550 ± 12, rank #614 of 1656 rated models, from 413 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| DoubtLess Evaluation - AIME25 and AMC23 - Concise prompt | 88.6 | Accuracy (%) | 100 |
| EuroEval Croatian NLU - Multi Wiki QA HR | 72.33 | Character F1 (%) | 99 |
| EuroEval Spanish NLU - MLQA ES | 69.84 | Character F1 (%) | 98 |
| EuroEval English NLU - ScaLA EN | 67.76 | Matthews correlation coefficient (%) | 97.3 |
| EuroEval Italian NLU - SQuAD IT | 75.65 | Character F1 (%) | 97.2 |
| EuroEval English NLU | 75.52 | NLU Average Score (%) | 97 |
| EuroEval Czech NLU - SQAD | 72.02 | Character F1 (%) | 96.9 |
| EuroEval Spanish NLU | 57.86 | NLU Average Score (%) | 94.9 |
| EuroEval Portuguese NLU | 63.84 | NLU Average Score (%) | 94.5 |
| EuroEval Spanish Summarization - Mlsum ES | 29.47 | chrF3++ (%) | 93.9 |
| EuroEval Bosnian NLU - Multi Wiki QA BS | 73 | Character F1 (%) | 93.8 |
| EuroEval English NLU - SST-5 | 69.3 | Matthews correlation coefficient (%) | 93.7 |
Interactive version: theaggregate.ai/model?slug=exaone-4-0-32b · How It Works · Data refreshed daily, snapshot 2026-10-20.