LG AI Research and Nvidia executives met on the afternoon of April 21 at LG AI Research headquarters in Magok, Seoul. They agreed to strengthen a technology alliance to expand the 'K-EXAONE' ecosystem.

Attendees included LG AI Research executives such as co-head of research Woo-hyung Lim (임우형) and EXAONE Lab head Jin-sik Lee (이진식). Nvidia executives also attended, including Bryan Catanzaro, vice president of applied deep learning research, who visited South Korea for a Nemotron developer conference, and So-young Jung (정소영), head of Nvidia Korea. They discussed strategies for cooperation on next-generation AI model development and for jointly building an AI ecosystem.

The companies agreed to expand the scope of cooperation, including jointly developing specialised, domain-focused models by combining LG's AI model EXAONE with Nvidia's Nemotron open ecosystem.

LG AI Research and Nvidia have continued technical cooperation in the development process from EXAONE 3.0 through K-EXAONE and the multimodal AI EXAONE 4.5 unveiled earlier this month.

South Korea ranked third, after the United States and China, with 5 noteworthy AI models in the AI Index Report recently published by Stanford University's Human-Centered AI institute. Of those, 4 were from LG AI Research's EXAONE series.

Nvidia Vice President Bryan Catanzaro said, "As a core partner of LG AI Research, Nvidia has joined forces to make EXAONE South Korea's top AI model." He said specialised models built with culture- and language-specific data unique to countries and industries have the potential to maximise AI possibilities. He stressed that combining LG EXAONE and Nvidia Nemotron would lead sovereign AI and contribute to expanding the ecosystem.

Woo-hyung Lim, co-head of research at LG AI Research, said, "Nvidia is a core technology partner that has worked with us on EXAONE development." He said LG and Nvidia would broaden their cooperation by a step into expanding the R&D ecosystem and produce sovereign AI outcomes that can be felt in industrial fields.