LG AI Research model product
K EXAONE 236B A23B
K-EXAONE is a state-of-the-art large language model developed by LG AI Research with 236 billion parameters.
Updated Aug 12, 2026. Default version: K-EXAONE-236B-A23B
LLMBoard score44.9K-EXAONE-236B-A23B
Coverage80%6 benchmark families
Context window32.8KTokens
Official input priceN/AOfficial price unavailable
K EXAONE 236B A23B Specifications
Technical details for the model's default version.
- Version
- K-EXAONE-236B-A23B
- Released
- Dec 31, 2025
- Knowledge cutoff
- Oct 1, 2025
- Parameters
- 236B
- Context window
- 32.8K
- Max output
- 32.8K
- Inputs
- text
- Outputs
- text
- Open weights
- No
- License
- Proprietary
K EXAONE 236B A23B Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
K EXAONE 236B A23B Arena Results
Preference and agent-evaluation results for the default version.
No Arena results
The default version does not have a matching Arena result yet.
K EXAONE 236B A23B Runtime Performance
Provider-specific output speed and catalog latency for K-EXAONE-236B-A23B. Runtime does not affect the capability score.
| FriendliAI | 50 tok/s | 0.4 s | 32.8K | 32.8K |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
K EXAONE 236B A23B Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- N/A
- Official provider
- N/A
- Lowest third-party
- N/A
- Tracked offerings
- 0
No provider prices
The default version has no current input or output token prices.
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
K EXAONE 236B A23B Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
| K-EXAONE-236B-A23B | 44.9 | 236B | 32.8K | 32.8K | No | Proprietary |
What is K EXAONE 236B A23B?
Key information about K EXAONE 236B A23B and its available data.
K-EXAONE is a state-of-the-art large language model developed by LG AI Research with 236 billion parameters. It features advanced reasoning capabilities, strong performance in both Korean and English, and excels at complex tasks including mathematics, coding, and creative writing.
The model uses A23B quantization for efficient inference while maintaining high quality outputs. K-EXAONE is designed for enterprise applications with robust instruction following and knowledge integration.
Data as of 2026-08-11.
K EXAONE 236B A23B vs nearby models
Open a comparison with the three ranked models immediately above and below this model.
Models similar to K EXAONE 236B A23B
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
OP
GPT-5-mini
OpenAI
NV
Nemotron 3 Super
NVIDIA
OP
o4 mini
OpenAI
SA
Sarvam 105B
Sarvam AI
AN
Claude Haiku 4.5
Anthropic
DE
DeepSeek-R1
DeepSeek
FAQ
Common questions about K EXAONE 236B A23B.
When was K EXAONE 236B A23B released?
K EXAONE 236B A23B's default version was released on Dec 31, 2025.
How much does K EXAONE 236B A23B cost?
No official standard PAYG price is currently available for K EXAONE 236B A23B.
Who created K EXAONE 236B A23B?
K EXAONE 236B A23B was created by LG AI Research.
What is the context window for K EXAONE 236B A23B?
The default version has a 32.8K token context window.
Is K EXAONE 236B A23B open weight?
No. The default version is not marked as having publicly available weights.
How many API providers offer K EXAONE 236B A23B?
No provider offering is currently linked to the default version.
What models should I compare K EXAONE 236B A23B with?
Nearby ranked alternatives include Claude Haiku 4.5, o4 mini, GPT-5-mini.