Skip to content
← Models
Open weightsLG AI Research/10 trustVerified

EXAONE 3.5 32B

LG's bilingual Korean-English flagship open model with 32K context.

Visit official page →

About

EXAONE 3.5 32B is the high-performance flagship of LG AI Research's EXAONE 3.5 family, released in December 2024 in three sizes (2.4B, 7.8B, 32B). It is a bilingual English-Korean instruction-tuned model with a 32K token context window, built on a decoder-only transformer with grouped-query attention. LG reports top-tier results among open models of similar size on practical usability, long-text comprehension, coding, and mathematics benchmarks.

Note the license caveat: EXAONE models are released under the EXAONE AI Model License Agreement, which restricts use to research and non-commercial purposes rather than a standard OSI-approved open-source license. The 32B model powers enterprise deployments such as ChatEXAONE inside LG.

Specs

Context window

Pricing

Free quota

API access

Weights

Open weights

Released

December 9, 2024

Benchmarks

Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.

Use cases

korean-nlpenterprise-raglong-document-qa

Trust score — 8/10

/10 trustVerified

Scored 8/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 115; Trending score 78/100 from researcher assessment.

More in the exaone family

Sources