Exaone 4.0 1.2B (Reasoning) — benchmark results
Provider: LG AI. Released 2025-07-15. Access: Open.
Unified ELO 1360 ± 43, rank #1381 of 1776 rated models, from 36 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA LiveCodeBench | 51.64 | Pass@1 (%) | 58.5 |
| AA AIME 2025 | 50.33 | Accuracy (%) | 48.5 |
| AA Humanity's Last Exam | 5.84 | Accuracy (%) | 42.9 |
| CritPt | 0 | Accuracy (self-reported) | 30.3 |
| AA GPQA Diamond | 51.52 | Accuracy (%) | 27.9 |
| AA CritPt | 0 | Accuracy (%) | 27.1 |
| AA MMLU-Pro | 58.77 | Accuracy (%) | 22.4 |
| AA Omniscience - Software Engineering (SWE) - Go | 8 | Accuracy (%) | 17.5 |
| AA TAU-2 Bench | 16.37 | Accuracy (%) | 14.6 |
| AA Omniscience - Software Engineering (SWE) - Java | 8 | Accuracy (%) | 12.3 |
| AA Omniscience - Software Engineering (SWE) - JavaScript | 15.45 | Accuracy (%) | 11.3 |
| AA SciCode | 9.26 | Accuracy (%) | 9.6 |
Interactive version: theaggregate.ai/model?slug=exaone-4-0-1-2b-reasoning · How the rankings work · Data refreshed daily, snapshot 2026-07-22.