Exaone 4.0 1.2B (Non-reasoning): benchmark results
Provider: LG AI. Released 2025-07-15. Access: Open.
Unified ELO 1369 ± 1, rank #1826 of 2032 rated models, from 84 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA Omniscience - Software Engineering (SWE) - Julia | 4.35 | Accuracy (%) | 39.4 |
| AA Humanity's Last Exam | 5.7 | Accuracy (%) | 35.5 |
| AA LiveCodeBench | 29.31 | Pass@1 (%) | 34.1 |
| AA Omniscience - Software Engineering (SWE) - Kotlin | 10.2 | Accuracy (%) | 27.8 |
| MERA Industrial - ruTXTMedQFundamental - Histology | 27.4 | Score (%) | 26.7 |
| MERA Industrial - ruTXTMedQFundamental - Pathological anatomy | 30.7 | Score (%) | 26.2 |
| AA AIME 2025 | 24 | Accuracy (%) | 25.3 |
| MERA Industrial - ruTXTMedQFundamental - General surgery | 31.9 | Score (%) | 23.8 |
| AA CritPt | 0 | Accuracy (%) | 23.2 |
| MERA Industrial - ruTXTMedQFundamental - Biophysics | 28.5 | Score (%) | 22.3 |
| MERA Industrial - ruTXTMedQFundamental - Pharmacology | 27.4 | Score (%) | 22.3 |
| MERA Industrial - ruTXTMedQFundamental - Clinical laboratory diagnostics | 28.1 | Score (%) | 21.8 |
Interactive version: theaggregate.ai/model?slug=exaone-4-0-1-2b-non-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-26.