K-EXAONE 2.0
LG AI Research
Key facts · verified Aug 15, 2026
K-EXAONE 2.0 is a LG AI Research model released on 31 Jul 2026. Pricing: Open weights, Apache 2.0. Its context window is 256k tokens. It is not yet ranked on LMArena's text leaderboard. Source: huggingface.co.
LG AI Research's K-EXAONE 2.0, released 31 July 2026 as Korea's largest AI foundation model to date: a mixture-of-experts model with 750B total and 37B active parameters, a 256k-token context window, and Apache 2.0 licensing that LG AI Research states allows commercial use without restrictions. LG AI Research's own model card reports 68.2% on SWE-Bench Verified and covers ten languages including Korean, English, Spanish and Japanese.
Pricing & specs checkedSource: vendor documentation · checked Aug 15, 2026
Pricing
Open weights, Apache 2.0
Context
256k tokens
LMArena Elo
Not rated
LMArena rank
Not ranked
Key Features
Open Weights Apache 2.0 Long Context MoE 37B Active / 750B Total Self-Hostable Multilingual (10 Languages)
Full Specifications
Open weights, Apache 2.0 | |
|---|---|
| Identity | |
| Developer | LG AI Research |
| Released | Jul 2026 |
| Status | GA |
| Licence | Open weights (Apache 2.0) |
| Self-hostable | Yes |
| Cost | |
| Input price | Not published by LG AI Research; K-EXAONE 2.0 is an open-weights release and per-token prices are set by the host serving it. |
| Output price | Not published by LG AI Research; K-EXAONE 2.0 is an open-weights release and per-token prices are set by the host serving it. |
| Cached input | $0.12 / 1M tokens (50%-off promo; $0.24 / 1M tokens regular) |
| Batch discount | Not offered |
| Free tier | Not offered |
| Capacity | |
| Context window | 256k tokens |
| Max output | 32k tokens |
| Long-context surcharge | None (flat FriendliAI rate across the 262k window) |
| Capability | |
| Vision in | No |
| Audio in | No |
| Function calling | Yes |
| Structured output | Not offered |
| Extended reasoning | Yes |
| Web search | Not offered |
| Code execution | Not offered |
| Access | |
| API | Yes |
| Chat app | No (k.exaone.ai demo currently offline) |
| Cloud marketplaces | Not offered |
| Fine-tuning | Not offered |