Korean LLM comparison

A.X 4.0 vs GPT

Independent attribute comparison. A paid diagnostic measures these on your own 20 prompts.

A.X 4.0GPT
VendorSK TelecomOpenAI
Params72B / 7B (경량)비공개
Context128K
LicenseOpen source · commercial OKAPI-only
PricingOpen weights, self-host / on-prem. Based on Qwen2.5 with added Korean training.Varies by tier; Korean token efficiency lower than Korea-tuned models (~-33% per A.X測).
Korean strengthKMMLU 78.3 (> GPT-4o 72.5); ~33% better Korean token efficiency than GPT-4o.Strong generalist; Korean cost-efficiency/honorific consistency can trail Korea-tuned models.
Best forOn-prem enterprise, Cost-sensitive Korean work, Data-sovereignty needsGeneral tasks, Ecosystem/tooling, Prototyping

How to decide

Decide on your own prompts in 48h
48h Diagnostic ₩490,000 · tax invoice available
Start paid diagnostic →

More comparisons