Connect with us

Technology

LG unveils New AI Model ‘EXAONE Deep’

Published

on

LG AI’s ‘EXAONE Deep’ marks a milestone in AI innovation, offering unparalleled reasoning capabilities and setting a new standard in the global AI landscape. 

SEOUL, South Korea, March 18, 2025 /PRNewswire/ — LG AI Research unveiled ‘EXAONE Deep,’ a globally focused reasoning AI model, as open source on the 18th. The release of ‘EXAONE Deep’ is significant as it proves the competitiveness of Korea’s own ‘Reasoning AI’ in the global AI market. LG AI Research introduces EXAONE Deep by participating in NVIDIA’s annual developer conference (GTC) held in San Jose, California, USA, from the 17th to the 21st local time.

Opening the Era of ‘Agentic AI’ by Developing Reasoning AI, Setting and Verifying Hypotheses Independently

LG AI Research announced the transition to the ‘Agentic AI’ era by releasing its own advanced reasoning model, ‘EXAONE Deep.’ Agentic AI refers to an active AI capable of making autonomous decisions by independently formulating hypotheses and conducting inferences to verify them. To implement this, the development of ‘Reasoning AI’ beyond the existing ‘Knowledge AI’ is essential. Currently, among the few global companies developing their own reasoning AI with foundation models, ‘EXAONE Deep’ is Korea’s first to achieve this milestone..

The 32B (32 billion parameter) Model Achieves World-Class Performance with a remarkably small parameter size

EXAONE Deep-32B (32 billion parameters) demonstrated exceptional efficiency and achieved remarkable performance, operating at just 5% of a very large competitor model.. LG AI Research released various performance evaluation results on the global open-source AI platform Hugging Face, assessing the logical reasoning skills and problem-solving capabilities of three ‘EXAONE Deep’ models with different parameters.

‘EXAONE Deep-32B’ proved its excellence in solving complex mathematical and scientific problems. In particular, ‘EXAONE Deep-32B’, based on the EXAONE foundation model with strengths in Korean, recorded the highest score of 94.5 in the 2025 CSAT mathematics area and achieved the top grade in all elective subjects (probability and statistics, calculus, geometry). Also, it recorded 95.7 points in MATH-500, an index evaluating mathematical problem-solving capability.

‘EXAONE Deep-32B’ also achieved excellent performance in indicators evaluating doctoral-level scientific reasoning capabilities in physics, chemistry, and biology. It recorded 66.1 points in the GPQA diamond test, which evaluates scientific problem-solving ability, outperforming competition with reasoning AI models of similar parameter scale, and also showed competitive performance with 59.5 points in LiveCodeBench, which evaluates coding ability.

LG AI Research expects that ‘EXAONE Deep’ will be highly utilized not only in professional fields required by industries in the future but also in scientific research and education fields such as physics and chemistry by showing high performance in evaluation indicators of specialized fields such as mathematics, science, and coding.

‘EXAONE Deep-32B’ was recognized for its technological competitiveness by being listed on the ‘Notable AI Models’ list selected by the US non-profit AI research organization Epoch AI with its open-source release.

LG’s AI Technology Maintains Overwhelming Performance While Significantly Reducing Model Size

LG AI Research also released the lightweight model ‘EXAONE Deep-7.8B’ and the on-device model ‘EXAONE Deep-2.4B’ developed with ‘EXAONE Deep-32B’ as open source. The lightweight model maintains up to 95% of the 32B’s performance despite being 24% of its size, and the on-device model has 86% of the performance despite being 7.5% of its size, possessing both high economic efficiency and practicality.

An official from LG AI Research said, “We announced EXAONE Deep about a month after attending the domestic AI industry competitiveness diagnosis and inspection meeting held at the National Artificial Intelligence Committee in February and foreshadowing the open-source release of a DeepSeek R1-level model soon,” and added, “The core of LG’s AI technology is maintaining performance while significantly reducing model size.”

In fact, the performance of the lightweight model 7.8B exceeded that of other leading lightweight models, and the 2.4B on-device model also showed world-class results in performance comparisons with equivalent models.

In particular, the on-device model has strengths in terms of security and personal information protection by safely processing data within the device without connecting to an external server. LG AI Research anticipates on-device models to rapidly spread in various industries such as smartphones, automobiles, and robots.

LG AI Research plans to lead the on-device AI market by enhancing models with LG affiliates such as LG Electronics and LG U+.

Meanwhile, Koo Kwang-mo, Chairman and CEO of LG Group emphasized in his New Year’s address this year, “The current LG has been built by accumulating many moments of challenging new areas and creating unprecedented value,” and added, “We will create a new lifestyle where people use cutting-edge technologies such as AI conveniently in their daily lives, allowing them to spend their precious time on more enjoyable and meaningful tasks.”

For more information, visit https://www.lgresearch.ai/blog/view?seq=543

About LG Group

LG Group is a leading global company representing South Korea, offering innovative products and services across various industries such as electronics, chemicals, telecommunications, and energy. Established in 1947, LG Group has grown into a world-renowned brand through its activities in these diverse fields. The company is committed to continuous research and development, focusing on innovation to enhance the quality of life for its customers. Emphasizing its role as a socially responsible enterprise, LG Group is striving to strengthen its competitiveness in the global market and achieve sustainable growth through its future portfolio in areas like AI, Bio, and Cleantech. The company is dedicated to realizing its vision of being a business that provides value to customers and society, pursuing this mission with unwavering determination.

About LG AI Research

Launched in December 2020 as the artificial intelligence (AI) research hub of South Korea’s LG Group, LG AI Research aims to lead the next epoch of artificial intelligence (AI) to realize a promising future by providing optimal research environments and leveraging state-of-the-art AI technologies. And LG AI Research developed its large-scale AI, EXAONE, a 300 billion parametric multimodal AI model, in 2021. EXAONE, which stands for “Expert AI for Everyone,” is a multi-modal large-scale AI model that stands out from its peers due to its ability to process both language and visual data. With one of the world’s largest learning data capacities, LG AI Research aims to engineer better business decisions through its state-of-the-art artificial intelligence technologies and its continuous effort on fundamental AI research. For more information, visit https://www.lgresearch.ai/.

 

 

 

View original content:https://www.prnewswire.com/news-releases/lg-unveils-new-ai-model-exaone-deep-302404567.html

SOURCE LG AI RESEARCH

Continue Reading
Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *

Technology

Trip.com Group Sincerely Accepts Administrative Penalty Decision Issued by the State Administration for Market Regulation of the People’s Republic of China

Published

on

By

SINGAPORE, July 25, 2026 /PRNewswire/ — Trip.com Group Limited (Nasdaq: TCOM; HKEX: 9961) today announced that it has received the administrative penalty decision issued by the State Administration for Market Regulation of the People’s Republic of China.

Trip.com Group sincerely accepts the decision and will adopt rectification measures in accordance with applicable laws and regulations to implement the decision’s requirements. The Company will strengthen its long-term governance mechanisms and strive to contribute to the sustainable development of the travel industry.

Trip.com Group’s management team will host a conference call at 8:00 AM U.S. Eastern Time on July 27, 2026 (or 8:00 PM Hong Kong Time on July 27, 2026).

The conference call will be available on Webcast live at: http://investors.trip.com.

All participants must pre-register to join this conference call using the participant registration link below:
https://register-conf.media-server.com/register/BIb78e08d8f18340c4882a7e4ab961906b.

Upon registration, each participant will receive details for this conference call, including dial-in numbers and a unique access PIN. To join the conference, please dial the number provided, enter your PIN, and you will join the conference instantly.

For further information, please contact:
Investor Relations
Trip.com Group Limited
Email: iremail@trip.com

View original content:https://www.prnewswire.com/news-releases/tripcom-group-sincerely-accepts-administrative-penalty-decision-issued-by-the-state-administration-for-market-regulation-of-the-peoples-republic-of-china-302834599.html

SOURCE Trip.com Group Limited

Continue Reading

Technology

NAVER Partners with Brookfield and NVIDIA to Expand Korea’s National AI Factory Infrastructure Buildout

Published

on

By

SAN FRANCISCO, July 25, 2026 /PRNewswire/ — NAVER, Brookfield and NVIDIA announced an expansion of Korea’s sovereign AI factory infrastructure. New investments will increase the initial NVIDIA DSX™ AI factory deployment from 55 megawatts to 200 megawatts.

Announced during Korea President Jae Myung Lee’s AI Summit visit to San Francisco, the planned 200-megawatt expansion will be built with the NVIDIA DSX platform at NAVER’s GAK Sejong hyperscale data center in Sejong, South Korea. The expanded infrastructure will provide Korea- and U.S.- based AI innovators with access to production-scale AI compute for building next-generation models, agents and AI-powered services.

Under the terms of the agreements, Brookfield will fund up to $9 billion as the exclusive capital partner, NVIDIA will invest $1 billion and NAVER will fund the remaining amount to finance the $10 billion project.

This builds on NAVER’s June announcement to extend its GAK Sejong data center with NVIDIA DSX, with a long-term path to gigawatt-scale sovereign AI infrastructure serving Korea’s enterprises, industries, government organizations and global AI cloud customers. Combining Brookfield’s capital with NVIDIA’s computing platform, the investment supports NAVER’s AI factory deployment.

“NVIDIA’s strategic investment and our infrastructure supply agreement with Brookfield have propelled NAVER’s vision for the AI Factory business into a robust execution phase,” said Haejin Lee, Founder and Chairman of NAVER. “Leveraging the solid partnerships with our global partners, we will drive technological innovation, foster a sovereign AI ecosystem, and spearhead efforts to strengthen South Korea’s AI competitiveness.” 

AI Factory Expansion and Open Model Collaboration to Fuel AI Innovators

NAVER, as an NVIDIA Cloud Partner, provides deep expertise in operating hyperscale infrastructure powered by the full-stack NVIDIA AI platform. The 200-megawatt AI factory, featuring NVIDIA Vera Rubin and Blackwell platforms, will establish a dedicated resource pool for emerging AI companies, providing the compute, software and support needed to develop and deploy competitive AI models and applications at scale.

This expanded infrastructure also builds on NAVER and NVIDIA’s collaboration on open model development for agentic and physical AI. NAVER is advancing its HyperCLOVA X models to be based on NVIDIA Nemotron™ 3 Ultra open models with its proprietary data and training expertise. NAVER is also the first Korean company to join the NVIDIA Nemotron Coalition, contributing to open model development across pretraining, post-training and reinforcement learning.

NAVER plans to launch an AI agent platform in Korea in the second half of the year, powered by NVIDIA Agent Toolkit software including NVIDIA NemoClaw™ blueprints. NAVER is also developing a Seoul World Model using proprietary urban street-view and spatial modeling data, built on NVIDIA Cosmos™ world foundation models.

About NAVER

Founded in 1999, NAVER is Korea’s largest Internet company and one of the world’s top tech companies. Leading cutting-edge technologies, NAVER operates the No.1 search engine in Korea and holds various business portfolios encompassing commerce, fintech, cloud, AI and robotics.

NAVER recorded sales of KRW 12.04 trillion (USD 8.18 billion) in 2025. TEAM NAVER continues to enhance its business portfolio and expand its global presence across Japan, North America, and Europe, while pursuing innovation through continuous research and development in future technologies.

View original content:https://www.prnewswire.com/news-releases/naver-partners-with-brookfield-and-nvidia-to-expand-koreas-national-ai-factory-infrastructure-buildout-302834577.html

SOURCE NAVER

Continue Reading

Technology

Beijing Review: Walking Through Time: China and U.S. Youths Explore Dali’s Past and Future

Published

on

By

BEIJING, July 24, 2026 /PRNewswire/ — On July 17, a China-U.S. youth delegation came to Dali of Yunnan Province. By examining how Dali’s rich history intersects with modern development, the delegates explored new pathways for rural development.

They visited Dali Old Town, tried their hand at making the Bai people’s Three-Course Tea and also explored the ancient town of Xizhou, where they learned how modern tourism and indigenous life coexist.

In Yunnanyi Village, they explored its history as a stop along the Tea Horse Road and learned about the role it played during the Second World War (WWII).

Tyler James Smith

“I think it’s a very underappreciated part of World War II history. Hearing these stories of different countries working together despite cultural differences is incredibly inspiring.

I also think there are many stories like these that haven’t been widely told, simply because World War II is such a complex period in history. That’s why I think it’s so meaningful to visit museums like this and experience these stories firsthand.”

At Xiangyun Economic and Technological Development Zone, they visited a local new energy company to see how green, low-carbon development is driving regional growth.

Valerie Marie

“I recently started studying energy transitions. I know China has been really big in the renewable energy sector. So actually getting to hear more about [China’s] 2060 [pledge], learning more about carbon neutrality [goals], as well as other zero-carbon goals, was cool.”

During their stay in Dali, they also strolled along the Erhai Lake Ecological Corridor.

Bai Yiwen

“I’d describe this journey as “to be continued,” because my own connection with Yunnan is far from over. For the U.S. delegates, this was only their first visit, so they’ve only had a glimpse of what Yunnan has to offer. I hope they will have more opportunities to come back to China, explore other cities in Yunnan, and discover even more of its people, culture and traditions.”

After Dali, the delegates will visit Beijing for more tours and exchanges. The event was co-hosted by China International Communications Group (CICG) Center for the Americas and the U.S.-based International Student Conferences.

https://x.com/beijingreview/status/2080104404552663067?s=46&t=yfVMVdMyE2zKAFrYaLoV-g

https://www.facebook.com/share/v/1EtGCCKzy4/?mibextid=wwXIfr

https://www.tiktok.com/@cachinachic/video/7665536913532587294?is_from_webapp=1&sender_device=pc&web_id=7227134149436605995

https://youtu.be/jGONWTqwduc?is=KfL-Zn3HyVYE-KEm

Contact: Jiaweibellapeng@163.com

View original content:https://www.prnewswire.com/news-releases/beijing-review-walking-through-time-china-and-us-youths-explore-dalis-past-and-future-302834587.html

SOURCE Beijing Review

Continue Reading

Trending