EXAONE 4.0: Unified Large Language Models Integrating Non-reasoning and Reasoning Modes
LG AI Research, Kyunghoon Bae, Eunbi Choi, Kibong Choi, Stanley Jungkyu Choi, Yemuk Choi, Kyubeen Han, Seokhee Hong, Junwon Hwang, Taewan Hwang, Joonwon Jang, Hyojin Jeon, Kijeong Jeon, Gerrard Jeongwon Jo, Hyunjik Jo, Jiyeon Jung, Euisoon Kim, Hyosang Kim, Jihoon Kim, Joonkee Kim, Seonghwan Kim, Soyeon Kim
2025-07-16
Summary
This paper talks about EXAONE 4.0, an advanced large language model that combines two modes: one for quick, easy answers and another for deep, careful thinking. It also supports using tools and speaks multiple languages including English, Korean, and Spanish.
What's the problem?
The problem is that language models before often specialized in either being fast and easy to use or in doing complex reasoning, but not both. Also, many models lacked good multilingual support and the ability to handle long, complicated tasks.
What's the solution?
EXAONE 4.0 solves this by integrating both fast non-reasoning and slow reasoning capabilities into one model, allowing it to switch between modes depending on the task. It adds the ability to use external tools like calculators or search engines, and it improves multilingual support by adding Spanish alongside English and Korean. The model design also allows it to handle very long text inputs efficiently.
Why it matters?
This matters because it brings together usability, deep reasoning, and tool use in one model, making it more powerful for a wide range of real-world applications. It helps AI understand and solve complex problems across languages and tasks, pushing forward what AI can do in ways closer to human thinking.
Abstract
EXAONE 4.0 integrates non-reasoning and reasoning modes, supports agentic tool use, and offers multilingual capabilities, demonstrating superior performance in its class.