IT Brief UK - Technology news for CIOs & IT decision-makers
Story image

LG AI Research & Google Cloud expand AI collaboration

Thu, 29th Aug 2024

LG AI Research and Google Cloud have announced an expansion of their collaboration to develop the latest family of EXAONE 3.0 generative AI models and the enterprise multimodal AI agent, ChatEXAONE.

This builds on earlier efforts between the firms to deliver innovative AI solutions for real-world industrial applications. In December 2021, the companies partnered to develop EXAONE 1.0, South Korea's first bilingual and multimodal generative AI model. This collaboration continued in July 2023 with EXAONE 2.0.

Under the latest agreement, LG's EXAONE 3.0 models and ChatEXAONE AI agent will be trained using Google Cloud's Cloud TPUs and GPUs. These resources are part of Google Cloud's AI Hypercomputer. The collaboration aims to overcome the significant computational demands required for such sophisticated models by utilising Google's advanced AI accelerators.

LG reported that leveraging the AI Hypercomputer resulted in a 56% reduction in inference processing time, a 35% decrease in memory usage, and a 72% cut in operating costs for the EXAONE 3.0 7.8B model compared to its predecessor. The EXAONE 3.0 model, which is bilingual in Korean and English, achieved world-class performance metrics in Korean.

"EXAONE 3.0 has been trained on 60 million pieces of specialised data sets, 350 million images, patents, software codes, mathematics, and chemistry. Google Cloud's AI Hypercomputer has been instrumental in achieving LG AI Research's goals of AI model lightweight technology and cost efficiency," said Hwayoung (Edward) Lee, Lead of AI Biz Dev and DX Group at LG AI Research.

Security considerations played a crucial role in LG's decision to partner with Google Cloud. LG AI Research, which works with affiliates that handle critical technologies such as large batteries, displays, and 5G, valued Google Cloud's built-in data governance, privacy controls, and robust AI data principles. Additionally, the data for training is based on LG's corporate databases and runs through Google Cloud's streamlined MLOps pipeline.

The collaboration also includes plans for LG to offer EXAONE 3.0 on the Google Cloud Marketplace. There will be efforts to develop APIs aimed at addressing the growing demand for enterprise AI agents.

"Our collaboration with LG AI Research exemplifies the power of partnership in pushing the boundaries of AI innovation. By providing LG with our cutting-edge AI infrastructure, we're not only accelerating their groundbreaking work but also shaping the future of industries worldwide," said Ki-sung Chi, Country Director of Google Cloud Korea.

Follow us on:
Follow us on LinkedIn Follow us on X
Share on:
Share on LinkedIn Share on X