Kanana / KoGPT family is a KR LLM from Kakao AI (South Korea).
One-line fit: best used for KakaoTalk-integrated AI.
Key specifications
| Provider | Kakao AI |
|---|---|
| Model | Kanana / KoGPT family |
| Type | KR LLM |
| Context window | Large |
| Pricing (indicative) | Product |
| Company category | Consumer KR |
| Region | South Korea |
| Official site | https://www.kakaocorp.com |
Benchmarks and public standing
Domestic KR
Treat scores as directional. SWE-bench, Terminal-Bench, LMSYS Arena, Artificial Analysis, and vendor cards use different harnesses and are not always comparable 1:1. Re-check the latest official model card before decisions.
What this model is good at
KakaoTalk-integrated AI
This aligns with Kakao AI’s broader strengths:
- KR messaging distribution
- Consumer AI features
Limitations and watch-outs
International API
- Rate limits, regional availability, and data retention policies vary by plan.
- Agent harness quality (Cursor, Claude Code, Codex, custom tools) can change outcomes more than raw model Elo.
- Open-weight availability (if any) is separate from hosted API quality and safety filters.
Ideal users
- Product engineers shipping features that match: KakaoTalk-integrated AI
- Teams standardizing on the Kakao AI ecosystem
- Agent builders who need a KR LLM
When to pick something else
- Cheaper volume: compare lower tiers from the same lab or open Chinese/EU alternatives.
- Maximum hard SWE thoughtfulness: compare Claude Fable-class and other coding flagships.
- Giant multimodal corpora / Workspace: compare Gemini-class models.
- Self-host / open weights: Llama, Qwen, DeepSeek, GLM, Mistral open lines.
Parent company
Kakao AI – Kakao ecosystem AI models for chat and services in Korea.
Related models from Kakao AI
Data is curated for AIForumSphere model directory mid-2026 directional figures.
