K-EXAONE

Closed weights LG AI Research 236B parameters December 2025

No estimate

No hardware requirements for this model

The weights for this model have not been published, so it cannot be downloaded or run on your own hardware at any size. It is reachable only through its provider, and no graphics card changes that.

On record

Full specification

Everything on record for this model. Most of it describes how it was trained rather than how it runs — useful context for judging how much work went into it, and how it compares with models built at a different scale.

Origin

Who built this model, where, and when it was published.

Organisation
LG AI Research
Organisation type
Industry
Country
Korea (Republic of)
Published
31 December 2025
Authors
Eunbi Choi, Kibong Choi, Seokhee Hong, Junwon Hwang, Hyojin Jeon, Hyunjik Jo, Joonkee Kim, Seonghwan Kim, Soyeon Kim, Sunkyoung Kim, Yireun Kim, Yongil Kim, Haeju Lee, Jinsik Lee, Kyungmin Lee, Sangha Park, Heuiyeen Yeen

What it does

The problem areas the model was built for. A model can carry several of each.

Domain
Language
Task
Language modeling/generation, Chat, Question answering, Code generation, Mathematical reasoning, Instruction interpretation

Size

How large the model is and how much data it was trained on. Parameters are the figure that decides whether it fits on a given graphics card.

Parameters
236B

236B

Training data
11,000,000,000,000 tokens

11T tokens

Epochs
1

Training compute

The arithmetic performed to train the model, measured in floating-point operations. It is a measure of what the training run cost, not of how fast the finished model answers you.

Training compute
1.5 × 10²⁴ FLOP

The training run

What it physically took to train: which chips, how many, for how long, and what that drew from the wall.

Training hardware
NVIDIA B200 GPUs
Chips used
512
Wall-clock time
3,240 hours (135 days)

K-EXAONE was trained using a cloud-based GPU environment. The total training duration was 135 days, utilizing 512 B200 GPUs. This corresponds to a total of 1,658,880 GPU hours, calculated as follows: 512 GPUs × 135 days × 24 hours = 1,658,880 GPU hours

Availability

Whether you can obtain the model and run it on your own hardware, which is what decides if any of the graphics-card figures on this page apply.

Weights
Closed — provider access only
Model access
API access
Training code
Unreleased

How it is classified

Labels the source dataset applies when tracking notable models, and how confident it is in the entry.

Why it is tracked
Training cost
Record confidence
Confident

Sources

Where this record came from and when it was last checked.

Reference
K-EXAONE Technical Report
Last updated
13 January 2026

What the numbers mean

Background

K-EXAONE was published by LG AI Research, in Korea (Republic of), in December 2025. The organisation is categorised as industry.

It works in Language, and is recorded as doing language modeling/generation, Chat, Question answering, Code generation, Mathematical reasoning, Instruction interpretation.

Its weights were never published, so it can only be reached through its provider. No graphics card changes that.

How it was trained

Training it took roughly 1.5 × 10²⁴ FLOP of computation, on NVIDIA B200 GPUs — a measure of what producing the model cost, not of how fast it answers.

It was trained on about 11,000,000,000,000 tokens of text.

Its inclusion criterion is training cost.

Answers

K-EXAONE — common questions

01

What GPU do I need to run K-EXAONE?

None. K-EXAONE is a closed model — its weights were never published, so it cannot be downloaded or run on your own hardware at any price. It is reachable only through its provider.

02

Is K-EXAONE open source?

No. K-EXAONE has not had its weights published, so it exists only as a service controlled by its owner.

03

How many parameters does K-EXAONE have?

K-EXAONE has 236B parameters. 236B. That figure is the total, and it is what decides how much memory the model needs — roughly half a gigabyte per billion at the compression most people use.

04

Who created K-EXAONE?

K-EXAONE was published by LG AI Research, based in Korea (Republic of), categorised as industry.

05

When was K-EXAONE released?

K-EXAONE was published in December 2025.

06

What is K-EXAONE used for?

K-EXAONE works in Language, and is recorded as handling language modeling/generation, Chat, Question answering, Code generation, Mathematical reasoning, Instruction interpretation. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

07

How much compute was used to train K-EXAONE?

Around 1.5 × 10²⁴ FLOP, on NVIDIA B200 GPUs. That measures what producing the model cost and says nothing about how quickly it answers once trained — inference speed comes from memory bandwidth, not from the training budget.

Source

Original publication

Record last updated 13 January 2026

The other direction

Looking at it from the other side?

This page starts from the model. If you already own a card and want to know everything it will run, start from the hardware instead.