News Report Technology

China Outpaces the USA with 130 LLMs in the Global AI Competition

China has surged to take the lead in this evolving technological scene, which is set to have a substantial impact throughout the 2020s. The swift expansion of Generative AI reliant on LLMs has sparked a metaphorical saying: 'When the wave recedes, we discover who wasn't prepared,' underscoring the urgency for national LLM frameworks.

As recorded in July this year, China has taken a significant leap ahead of the United States, showcasing an impressive number 130 LLMs within its territory. This lead emphasizes China’s ambition to be the leading global force in deploying LLMs across various fields, particularly in industrial sectors, reinforcing its status as the technological powerhouse of the world.

As things stand, the uptake of LLMs within the industrial sector is relatively limited, a fact that falls short of China's lofty aspirations. The leadership is actively working to rectify this deficiency in the near future.

There's a common misunderstanding that businesses merely need AIs that can outperform humans in conversation, but the core need is actually for solutions that provide genuine, measurable advantages, yielding substantial profit boosts. Achieving this goal necessitates the establishment of a comprehensive national LLM structure, enabling companies to develop tailored LLMs for unique uses. LLMs Leading this ambitious project is

the Qianfan large model platform, which is set to be the bedrock of China's forthcoming national LLM structure. Qianfan aims to assist businesses in quickly creating customized LLMs that address their specific demands, offering access to a selection of 42 major LLMs from both domestic and international research centers, including the Baidu’s In the coming months, industry bodies will be crucial in determining whether LLMs will truly become the intelligent operating system for this AI-driven age. As the wave of LLM adoption reaches its peak around next summer, China hopes to emerge victorious, glancing with a smirk at those nations that may have been ‘caught unprepared’ by not setting up a solid national LLM framework amidst the ongoing frenzy. open-source Meta model Llama 2 .

Recently, the prominent Chinese technology company, owner of WeChat, disclosed that

  • has now become fundamental to over 50 of its available products and services. According to Tencent's vice president, Jiang Jie, by July there will be over 130 large language models operational in China. The introduction of Hunyuan follows the recent release of AI models by several Chinese tech companies, including Baidu and SenseTime Group. Hunyuan boasts over 100 billion parameters and was trained with more than 2 trillion tokens—two significant indicators of AI models' capabilities. Tencent asserts that its model surpasses OpenAI's ChatGPT in various aspects, including generating lengthy texts containing thousands of words and addressing specific mathematical challenges. Moreover, it reportedly experiences 30% fewer ‘hallucinations’ compared to Llama 2. The term ‘hallucination’ refers to instances when AI models produce incorrect information yet present it as authoritative. Hunyuan 8 Essential Insights Regarding Large Language Models

The article was written with the  Telegram channel ‘s assistance.

Read more about AI:

Disclaimer

In line with the Trust Project guidelines Damir plays a key role as the team leader, product manager, and editor at Metaverse Post, focusing on topics like AI/ML, AGI, LLMs, Metaverse, and Web3-related fields. His articles draw in an audience of over a million users per month. With a decade of experience in SEO and digital marketing, he’s been recognized by many reputable publications such as Mashable, Wired, Cointelegraph, The New Yorker, Inside.com, Entrepreneur, BeInCrypto, among others. Damir traverses between the UAE, Turkey, Russia, and the CIS as a digital nomad. With a bachelor's degree in physics, he credits this background for equipping him with the critical thinking abilities essential for thriving in the constantly evolving landscape of the internet. 

  • Press Releases
  • by