LG AI Research

K-EXAONE 2.0

LG AI Research

K-EXAONE 2.0 is LG AI Research’s 750B-parameter sovereign language model with 37B active parameters. Its hybrid-attention mixture-of-experts architecture scales the preceding K-EXAONE release while expanding support to ten languages, including Korean, English, Japanese, and several European languages. Apache 2.0 replaces the earlier generation’s custom licensing, allowing commercial weight deployment. The model is designed for reasoning, coding, and agentic applications with 256K context. Official weights, a technical report, and demo access support reproducible evaluation rather than relying only on the provider’s aggregate benchmark claims.

EXAONE 4.5

LG AI Research

EXAONE 4.5 is LG AI Research’s first open-weight vision-language model, combining a 33B total parameter architecture with a 1.2B vision encoder. It extends the EXAONE family into image understanding, technical documents, charts, and multilingual visual reasoning, with a 256K context window. The release publishes deployment code and evaluation information alongside weights. Its custom EXAONE license must be consulted for usage rights rather than treating the model as Apache-licensed. It is suited to research and domain-specific document assistants, while the actual reliability of extracted facts and visual reasoning requires application-level validation.

K-EXAONE

LG AI Research

K-EXAONE is LG’s first large sparse sovereign language release, designed for Korean and multilingual reasoning, coding, and agents. It combines a 236B capacity with 23B active parameters and a 256K context using hybrid attention. Alignment incorporates Korean cultural and historical context. Public weights and deployment instructions support private evaluation, but the custom license governs commercial use. Benchmark comparisons should preserve checkpoint size, thinking mode, and evaluation protocol.

EXAONE 4.0

LG AI Research

EXAONE 4.0 is LG’s hybrid reasoning family with ordinary and thinking modes. The 32B checkpoint combines local and global attention, while the 1.2B tier targets lower-resource use. It supports Korean, English, and additional languages, tool calls, and long-context understanding. Public weights and deployment instructions support private evaluation, but the custom license governs commercial use. Benchmark comparisons should preserve checkpoint size, thinking mode, and evaluation protocol.

EXAONE Deep

LG AI Research

EXAONE Deep is LG’s reasoning-specialized language family for mathematics, coding, and scientific problems. It groups 2.4B, 7.8B, and 32B checkpoints with 32K context. Reasoning traces improve difficult task performance while adding inference cost. Public weights and deployment instructions support private evaluation, but the custom license governs commercial use. Benchmark comparisons should preserve checkpoint size, thinking mode, and evaluation protocol.