More Than US, China Launched 79 Major Language Models In Last 3 Years

China has launched 79 large language models (Large-Language Models/LLMs) (photo: doc. pexels)

JAKARTA - Several organizations in China have launched 79 large-Language Models (LLMs) in the country over the past three years in an effort to strengthen the development of artificial intelligence (AI) algorithms. This was reported by a research institute run by the country's Ministry of Science and Technology.

The development of LLMs, which are trained using deep learning techniques on large amounts of text data, entered an "acceleration" phase in 2020, according to a report authored by a research institute run by China's Ministry of Science and Technology.

In 2020, Chinese organizations launched 2 LLMs, compared to 11 launched in the United States. However, in 2021, there will be a total of 30 LLMs launched in each country, according to the report published on Sunday, 28 May.

United States organizations overall launched 37 LLMs the following year, while China launched 28, according to data compiled in this report, some of whose authors include the China Scientific and Technical Information Institute.

To date this year, China leads the way with 19 LLMs compared to 18 LLMs in the United States.

"If you look at the distribution of major language models rolled out worldwide, China and the United States are ahead by large margins, covering more than 80% of the global total," the report concludes, as cited in a press release summarizing the report's main findings. "The United States has always been ranked first in the world in terms of the number of major language models."

The report comes as China's artificial intelligence industry faces significant challenges due to US-led export curbs that prevent Chinese organizations from accessing semiconductors used to train LLMs, among other advanced computing tasks.

The report analyzes 79 LLMs developed in China, noting that although there are already 14 provinces and regions where the technology is developed, co-development projects between academia and industry are still "insufficient".

After OpenAI launched ChatGPT, Chinese tech giants, ranging from Alibaba to surveillance company Sensetime and search engine giant Baidu, have launched their own versions of chatbots powered by generative artificial intelligence and LLMs.