{"id":41409,"date":"2025-07-28T01:03:10","date_gmt":"2025-07-28T01:03:10","guid":{"rendered":"https:\/\/www.newsbeep.com\/us\/41409\/"},"modified":"2025-07-28T01:03:10","modified_gmt":"2025-07-28T01:03:10","slug":"huawei-launches-cloudmatrix-384-server-as-an-alternative-to-nvidias-ai-infrastructure-stack","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/us\/41409\/","title":{"rendered":"Huawei launches CloudMatrix 384 server as an alternative to Nvidia’s AI infrastructure stack"},"content":{"rendered":"

China\u2019s Huawei Technologies Co. Ltd.<\/a> threw down the gauntlet to Nvidia Corp. Saturday when it revealed its most powerful artificial intelligence server system to date. The company unveiled the CloudMatrix 384 system at the World Artificial Intelligence Conference in Shanghai, where dozens of local companies showed off their latest AI hardware.<\/p>\n

Reuters reported<\/a> that Huawei is positioning the new CloudMatrix system as a direct rival to Nvidia\u2019s premium server product, the GB200 NVL72, which is subject to export controls and cannot be sold in China.<\/p>\n

CloudMatrix 384 is powered by 384 of Huawei\u2019s most sophisticated AI chips, the Ascend 910C graphics processing unit. That compares with just 72 B200 GPUs found in Nvidia\u2019s GB200 NVL72 system.<\/p>\n

The design of the CloudMatrix system illustrates Huawei\u2019s strategy for taking on Nvidia in the AI industry. Though Nvidia\u2019s GPUs are more powerful on an individual level, Huawei believes it can overcome this deficiency by stacking more of its chips into a single cluster, allowing them to work together to overcome their lower performance. Huawei terms the idea as a \u201csupernode\u201d chip architecture, and says it has created high-speed network interconnection technology to enable the chips to communicate with each other.<\/p>\n

Huawei makes some stunning claims, saying that the CloudMatrix 384 delivers around 300 petaflops<\/a> of compute performance, surpassing the 180-petaflop limit of Nvidia\u2019s GB200 NVL72 system. It also beats Nvidia\u2019s product in terms of memory and bandwidth, although the tradeoff is that it uses significantly more energy to do so. According to Huawei, the CloudMatrix 384 consumes 559 kilowatts per hour, which means it\u2019s almost four times as power-hungry as Nvidia\u2019s system.<\/p>\n

Nonetheless, it\u2019s clear that Huawei has decided that sheer computing power is what counts most of all, and with the launch of the CloudMatrix 384, it can now offer an alternative to Chinese firms that have been blocked from accessing Nvidia\u2019s most advanced GPUs due to U.S. sanctions.<\/p>\n

Huawei first revealed the CloudMatrix 384 system in a low-key announcement in April, but it attracted little fanfare then because it wasn\u2019t yet available to customers. However, some analysts did take notice of it\/ Dylan Patel from SemiAnalysis said in a blog post<\/a>\u00a0that Huawei \u201cnow has AI system capabilities that could beat Nvidia.\u201d<\/p>\n

Praising Huawei\u2019s networked clustering capabilities, Dylan said the CloudMatrix 384 surpasses the performance of Nvidia\u2019s flagship server rack on a number of key metrics, and said it will become the obvious choice of infrastructure for Chinese AI companies.<\/p>\n

Nvidia Chief Executive Jensen Huang said something similar in May, telling Bloomberg<\/a> that Huawei has been \u201cmoving quite fast,\u201d and highlighting the CloudMatrix 384 system as an example of that. Around the same time as the CloudMatrix 384 was announced, Huawei also unveiled an even more advanced processor called the Ascend 920, although that model has yet to go on sale.<\/p>\n

Huawei threatens to corner domestic AI market<\/p>\n

Huang\u2019s comments came just one month after U.S. President Donald Trump ramped up export controls<\/a> on Nvidia\u2019s chip exports, banning the company from shipping its H20 GPUs to China \u2013 a lower-powered version of the H200 GPU that had been specifically designed to meet earlier prohibitions on exports to China.<\/p>\n

Trump\u2019s decision effectively ended all Nvidia chip sales to China, but in a stunning turnaround, the White House reversed its course on July 15, saying it would grant special licenses<\/a> to Chinese firms that want to import the H20, reopening the market for Nvidia.<\/p>\n

At the time, the White House\u2019s AI Czar David Sacks said the U-turn was justified, because it will help Nvidia prevent Huawei from cornering the domestic market and generating more revenue that would be thrown at advanced chip research.<\/p>\n

However, that narrative does not tell the full story, for Commerce Secretary Howard Lutnick told Reuters<\/a> that the decision to grant export licenses was made in connection with a broader trade deal that enables the U.S. to import vital rare earth minerals from China.<\/p>\n

In addition, there are question marks about the effectiveness of the U.S. sanctions on China. Last week, Nvidia told CNBC that building data centers with smuggled chips would be a \u201closing proposition\u201d from a technical and financial perspective, in response to reports that Chinese AI companies are acquiring them from the black market.<\/p>\n

Earlier, a Financial Times report revealed that over $1 billion worth of advanced Nvidia chips had been smuggled into China<\/a> via unofficial channels since May, including the B200 GPUs.<\/p>\n

Photo: Huawei<\/p>\n

Support our open free content by sharing and engaging with our content and community.<\/p>\n

Join theCUBE\u00a0Alumni Trust Network<\/p>\n

Where Technology Leaders Connect, Share Intelligence & Create Opportunities<\/p>\n

11.4k+\u00a0\u00a0<\/p>\n

CUBE Alumni Network<\/p>\n

C-level and Technical<\/p>\n

Domain Experts<\/p>\n

Connect with 11,413+ industry leaders from our network of tech and business leaders forming a unique trusted network effect.<\/p>\n

SiliconANGLE Media is a recognized leader in digital media innovation serving innovative audiences and brands, bringing together cutting-edge technology, influential content, strategic insights and real-time audience engagement. As the parent company of SiliconANGLE<\/a>, theCUBE Network<\/a>, theCUBE Research<\/a>, CUBE365<\/a>, theCUBE AI<\/a> and theCUBE SuperStudios \u2014 such as those established in Silicon Valley and the New York Stock Exchange (NYSE) \u2014 SiliconANGLE Media operates at the intersection of media, technology, and AI. . <\/p>\n

Founded by tech visionaries John Furrier and Dave Vellante, SiliconANGLE Media has built a powerful ecosystem of industry-leading digital media brands, with a reach of 15+ million elite tech professionals. The company\u2019s new, proprietary theCUBE AI Video cloud is breaking ground in audience interaction, leveraging theCUBEai.com neural network to help technology companies make data-driven decisions and stay at the forefront of industry conversations.<\/p>\n","protected":false},"excerpt":{"rendered":"China\u2019s Huawei Technologies Co. Ltd. threw down the gauntlet to Nvidia Corp. Saturday when it revealed its most…\n","protected":false},"author":2,"featured_media":41410,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[45],"tags":[182,181,507,33338,29010,15091,74],"class_list":["post-41409","post","type-post","status-publish","format-standard","has-post-thumbnail","category-artificial-intelligence","tag-ai","tag-artificial-intelligence","tag-artificialintelligence","tag-huawei-launches-cloudmatrix-384-server-as-an-alternative-to-nvidias-ai-infrastructure-stack","tag-mike-wheatley","tag-siliconangle","tag-technology"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/posts\/41409","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/comments?post=41409"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/posts\/41409\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/media\/41410"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/media?parent=41409"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/categories?post=41409"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/tags?post=41409"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}