Exaone 4.0 1.2B (Non-reasoning)
Open weightsOpen weightsText1.2B parametersDirect answers
EXAONE family model by LG AI Research — non-reasoning.
About the EXAONE family
LG AI Research's EXAONE models (1.2B–33B, plus the K-EXAONE Korean sovereign flagship) — bilingual Korean/English models with reasoning toggles from one of Korea's biggest conglomerates.
All EXAONE modelsHow this one is configured
This configuration runs in non-reasoning mode (direct answers, faster and cheaper); ~1.2B parameters.
Who makes it
LG AI Research — South Korea. Not listed as a provider in this directory.