Nota (CEO Myung-soo Chae), a company specializing in AI model lightweighting and optimization, is accelerating the spread of LLM by applying Nota's AI lightweighting technology to LG AI Research Institute's EXAONE.

Nota announced on the 10th that it has signed a strategic partnership agreement with LG AI Research Center for the commercialization of the next-generation large language model (LLM), EXAONE.

This collaboration includes: △creating synergy by applying Nota's AI lightweight technology to ExaOne; △promoting joint business cooperation through Nota's solutions; and the two companies aim to expand the AI market and accelerate its application in industrial sites through close cooperation.

With this partnership, the two companies have established a concrete cooperation structure based on ExaOne.

LG AI Research Center has improved the efficiency of its technical support by utilizing ExaOne in a wide range of industrial fields through Nota's solution, and Nota has secured a new business opportunity by supporting ExaOne in a large number of devices and utilizing it to supply its own solutions.

Nota CEO Cha Myung-soo said, “This partnership is very meaningful in that it is a win-win structure that allows both companies to strengthen their competitiveness in the AI solution market and grow together,” and added, “We will actively support LG AI Research Institute’s high-performance ExaOne model so that it can be applied to various industries.”

ExaOne is a next-generation large-scale language model developed by LG AI Research Institute., can simultaneously process various types of data, including natural language and images. It features high-performance inference capabilities and language understanding and generation functions, and is characterized by its wide range of applications, from server to on-device environments.

Nota's AI lightweight technology reduces the computational load and memory usage of large-scale AI models, such as Exa1, enabling them to operate efficiently in various semiconductor environments.

Key techniques include pruning, which removes unnecessary operations, and quantization, which reduces the number of weight bits. These techniques will enable Exa1 to be efficiently utilized in diverse industrial environments, including transportation and industrial safety.

“The partnership with Nota is a significant opportunity to expand ExaOne’s innovations to a wider range of industries,” said Lee Hwa-young, Executive Director of AI Business Development at LG AI Research Center. “We expect it to significantly contribute to creating tangible business value.”