LG's K-EXAONE 2.5 API is free to use, all you need is an API key
Free K-EXAONE 2.5 API access with your own API key
LG AI Research has made the K-EXAONE 2.5 API free. Developers, companies and researchers can create an API key on the EXAONE API Platform and use this 512K-context agent model at no cost.
- Eligible plans
- K-EXAONE 2.5 API
- Eligibility
- Open to developers, companies and research institutions; sign-up currently available in selected countries and regions
The briefing
On November 9, LG AI Research made the API for its in-house model K-EXAONE 2.5 free to use. Developers, startups, companies and research institutions can grab an API key on the EXAONE API Platform and use the model in their own services or research at no cost.
What you can use
K-EXAONE 2.5 is a further-trained version of the August release. It has 750B total parameters with 31B active, a 512K context window, Korean and English support, and handles chat, tool calling, structured outputs, code generation and streaming.
It is built for agent work. LG reports 59.2 on SWE-Bench Pro, 78.0 on BrowseComp, 74.6 on AutomationBench and 77.3 on Claw-Eval, putting it in the same range as DeepSeek-V4.1-Flash and GLM-5.3.
The model always reasons before answering. You can set reasoning_effort to low, high or max, and low gives the fastest responses.
Sign-up is currently open in selected countries and regions only, and requests are rate-limited. If you hit the limit you'll get a 429 and should retry after a short wait.
How to use
- Open the EXAONE API Platform, sign up and create an API key.
- The API is OpenAI-compatible. Point your client at https://api.exaone.ai/v1/chat/completions with the model name k-exaone_v2.5, and existing OpenAI clients or agent tools work after swapping the base URL and key.
Details: LG AI Research announcement and K-EXAONE 2.5 API docs.
