K-EXAONE (Non-reasoning) — benchmark results

K-EXAONE evaluated with reasoning disabled. Provider: LG AI. Released 2026-01-12. Access: Open.

Unified ELO 1513 ± 21, rank #758 of 1776 rated models, from 52 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
AA MMLU-Pro80.98Accuracy (%)73
AA-LCR47Score (self-reported)62.1
AA Long Context Reasoning47Accuracy (%)56.3
AA TAU-2 Bench59.06Accuracy (%)56.2
Artificial Analysis Intelligence Index16.74Intelligence Index52.6
AA Omniscience - Software Engineering (SWE) - R12Accuracy (%)52
AA GPQA Diamond69.49Accuracy (%)51.7
AA Omniscience - Software Engineering (SWE) - Swift36Accuracy (%)47.7
AA AIME 202544Accuracy (%)44.4
AA Humanity's Last Exam5.38Accuracy (%)39.7
AA IFBench39.59Accuracy (%)37.6
AA Global-MMLU-Lite - German85.75Accuracy (%)37.3

Interactive version: theaggregate.ai/model?slug=k-exaone-non-reasoning · How the rankings work · Data refreshed daily, snapshot 2026-07-22.