Alibaba releases AI model it says rivals DeepSeek, OpenAI, Meta’s top models

MT HANNACH
4 Min Read
Disclosure: This website may contain affiliate links, which means I may earn a commission if you click on the link and make a purchase. I only recommend products or services that I personally use and believe will add value to my readers. Your support is appreciated!

The Chinese giant of Alibaba technology flexes its muscles in the world race to Artificial Intelligence (AI) Dominance, saying that the latest version of its Qwen 2.5 can take the best rivals of both foreign and national.

Do not be outstanding the attention surrounding the Chinese company Deepseek – which rocked the American technology sector this week with its alleged emergence as a potential rival for the main American IA – Alibaba companies says that its last offer is superior to them all in almost all measures.

The Alibaba group logo in its Beijing office

Alibaba said the latest version of his Qwen 2.5 can take the best rivals of both foreign and national. (Photos Thomas Peter / Reuters)

Teleprinter Security Last Change Change %
Baba Alibaba Holding Ltd. 97.97 +1.90

+ 1.98%

Alibaba Holding Group Ltd – ADR

“Qwen 2.5-Max surpasses … Almost all the GPT-4O, Deepseek-V3 and Llama-3.1-405b plates,” said the Alibaba Cloud unit in an Official WeChat account, referring, referring To open And the Meta’s most advanced Open-Source IA models.

The technology magnate doubts Deepseek claims, says that the American media fell into “CCP propaganda”

Alibaba’s Qwen account on X displayed statistics indicating how the model accumulates against its competitors, saying: “He obtains competitive performance against first -level models and outperforms Deepseek V3 in benchmarks like Arena Hard, Livebench, Livecodebench, GPQA-DIAMOND. “

The unusual timing of the release of Qwen 2.5-Max, the first day of the Lunar New Year when most Chinese are out of work and with their families, underlines the pressure that the Chinese startup of AI Deepseek Rise in the last three Weeks have not only rivals abroad, but also its internal competition.

Trump, CEO of Openai Pasa on Deepseek Frenzy

The January 10 release of the AI ​​assistant from Deepseek, fueled by the Deepseek-V3 model, as well as the January 20 output of its R1 model, shocked the Silicon Valley and caused a plunge of technological sharesWith the alleged development and use costs of the Chinese startup, encouraging investors to question huge spending plans by directing AI companies in the United States

But Deepseek’s success has also led to a race among its national competitors to upgrade their own AI models.

Get Fox Affairs on the move by clicking here

Two days after the release of Deepseek-R1, Tiktok Owner bytedance has published an update of its Lighthouse AI model, which affirmed O1 of Openai supported by Microsoft in Aime, a reference test which measures the way in which AI models include and react to complex instructions.

This has echoed the assertion of Deepseek that his R1 model competed with O1 of Openai on several performance marks.

Reuters contributed to this report.

Share This Article
Leave a Comment

Leave a Reply

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