
Artificial Analysis' Intelligence Index / Courtesy of LG AI Research
EXAONE 4.0, the latest artificial intelligence (AI) system developed by LG AI Research, was ranked the world’s 11th most intelligent model, according to a new analysis released Wednesday.
The ranking was based on the Intelligence Index by global AI performance evaluation agency Artificial Analysis.
In the rankings, which use a composite metric based on seven performance indicators, EXAONE 4.0 32B scored 64.09 points, trailing behind models including Anthropic’s Claude 4 Opus, Google’s Gemini 2.5 Flash, Alibaba’s Qwen3 235B, OpenAI’s o3-pro and xAI’s Grok 4.
Among publicly available models, EXAONE 4.0 placed fourth, following Qwen3, DeepSeek R1 and GLM-4.5.
With EXAONE 4.0, LG AI Research became the world’s eighth competitive AI developer, following global titans such as xAI, OpenAI, Google, Alibaba and DeepSeek. LG was also the only Korean company to make it into the top 10.
On July 15, LG AI Research released the open-weight version of EXAONE 4.0 on Hugging Face for research, academic and educational purposes. The 32B model, which has 32 billion parameters, surpassed 500,000 downloads just two weeks after its release, the fastest phase for a Korean AI model.
Reasoning models EXAONE Deep and EXAONE 4.0 were previously both included in the “Notable AI Models” list by Epoch AI, a U.S.-based nonprofit AI research institute.
The list is used as a reference for comparing national and corporate AI competitiveness in the annual AI Index Report published by Stanford University.
“The latest rankings show that LG EXAONE has the capability and potential to compete with the world’s leading frontier AI models,” LG AI Research co-head Lee Hong-lak said.
“We will continue striving to develop world-class AI models to help Korea become one of the top three global AI powerhouses.”
LG Group has been placing its strategic focus on its AI business, with Chairman Koo Kwang-mo highlighting AI, bio and clean tech as the group’s top business priorities.