{"id":8972,"date":"2025-08-19T08:48:08","date_gmt":"2025-08-19T08:48:08","guid":{"rendered":"https:\/\/www.europesays.com\/ie\/8972\/"},"modified":"2025-08-19T08:48:08","modified_gmt":"2025-08-19T08:48:08","slug":"nvidia-working-on-new-ai-chip-for-china-that-outperforms-the-h20-reuters-reports","status":"publish","type":"post","link":"https:\/\/www.europesays.com\/ie\/8972\/","title":{"rendered":"Nvidia working on new AI chip for China that outperforms the H20, Reuters reports"},"content":{"rendered":"

Nvidia<\/a>\u00a0is developing a\u00a0new\u00a0AI\u00a0chip\u00a0for\u00a0China\u00a0based on its latest Blackwell architecture that will be more powerful than the\u00a0H20\u00a0model it is currently allowed to sell there, two people briefed on the matter said.<\/p>\n

U.S. President Donald Trump last week\u00a0opened the door\u00a0to the possibility of more advanced\u00a0Nvidia\u00a0chips being sold in\u00a0China. But the\u00a0sources\u00a0noted U.S. regulatory approval is far from guaranteed amid deep-seated fears in Washington about giving\u00a0China\u00a0too much access to U.S. artificial intelligence technology.<\/p>\n

The\u00a0new\u00a0chip, tentatively known as the B30A, will use a single-die design that is likely to deliver half the raw computing power of the more sophisticated dual-die configuration in\u00a0Nvidia’s flagship B300 accelerator card, the\u00a0sources\u00a0said.<\/p>\n

A single-die design is when all the main parts of an integrated circuit are made on one continuous piece of silicon rather than split across multiple dies.<\/p>\n

The\u00a0new\u00a0chip\u00a0would have high-bandwidth memory and\u00a0Nvidia’s NVLink technology for fast data transmission between processors, features that are also in the\u00a0H20\u00a0– a\u00a0chip\u00a0based on the company’s older Hopper architecture.<\/p>\n

The\u00a0chip’s specifications are not completely finalized but\u00a0Nvidia\u00a0hopes to deliver samples to Chinese clients for testing as early as next month, said the\u00a0sources\u00a0who were not authorized to speak to media and declined to be identified.<\/p>\n

Nvidia\u00a0said in a statement: “We evaluate a variety of products for our roadmap, so that we can be prepared to compete to the extent that governments allow.”<\/p>\n

“Everything we offer is with the full approval of the applicable authorities and designed solely for beneficial commercial use,” it said.<\/p>\n

The U.S. Department of Commerce did not respond to a Reuters request for comment.<\/p>\n

Flashpoint<\/p>\n

The extent to which\u00a0China, which generated 13% of\u00a0Nvidia’s revenue in the past financial year, can have access to cutting-edge\u00a0AI\u00a0chips is one of the biggest flashpoints in U.S.-Sino trade tensions.<\/p>\n

Nvidia\u00a0only received permission in July to recommence sales of the\u00a0H20. It was developed specifically for\u00a0China\u00a0after export restrictions were put in place in 2023, but the company was abruptly ordered to stop sales in April.<\/p>\n

Trump said last week he might allow\u00a0Nvidia\u00a0to sell a scaled-down version of its next-generation\u00a0chip\u00a0in\u00a0China\u00a0after announcing an unprecedented deal that will see\u00a0Nvidia\u00a0and rival AMD<\/a>\u00a0give the U.S. government 15% of revenue from sales of some advanced\u00a0chips in\u00a0China.<\/p>\n

A\u00a0new\u00a0Nvidia\u00a0chip\u00a0for\u00a0China\u00a0might have “30% to 50% off”, he suggested in an apparent reference to the\u00a0chip’s computing power, adding that the\u00a0H20\u00a0was “obsolete”.<\/p>\n

U.S. legislators, both Democratic and Republican, have worried that access to even scaled-down versions of flagship\u00a0AI\u00a0chips will impede U.S. efforts to maintain its lead in artificial intelligence.<\/p>\n

But\u00a0Nvidia\u00a0and others argue that it is important to retain Chinese interest in its\u00a0chips \u2014 which work with\u00a0Nvidia’s software tools \u2014 so that developers do not completely switch over to offerings from rivals like Huawei.<\/p>\n

Huawei has made great strides in\u00a0chip\u00a0development, with its latest models said to be on par with\u00a0Nvidia\u00a0in some aspects like computing power, though analysts\u00a0say\u00a0it lags in key areas such as software ecosystem support and memory bandwidth capabilities.<\/p>\n

Complicating\u00a0Nvidia’s efforts to retain market share in\u00a0China, Chinese state media have also in recent weeks alleged that the U.S firm’s\u00a0chips could\u00a0pose security risks, and authorities\u00a0have cautioned\u00a0Chinese tech firms about purchasing the\u00a0H20.\u00a0Nvidia\u00a0says its\u00a0chips carry no backdoor risks.<\/p>\n

Nvidia\u00a0is also preparing to start delivering a separate\u00a0new\u00a0China-specific\u00a0chip\u00a0based on its Blackwell architecture and designed primarily for\u00a0AI\u00a0inference tasks, according to two other people familiar with those plans.<\/p>\n

Reuters\u00a0reported in May\u00a0that this\u00a0chip, currently dubbed the RTX6000D, will sell for less than the\u00a0H20, reflecting weaker specifications and simpler manufacturing requirements.<\/p>\n

The\u00a0chip\u00a0is designed to fall under thresholds set by the U.S. government. It uses conventional GDDR memory and features memory bandwidth of 1,398 gigabytes per second, just below the 1.4 terabyte threshold established by restrictions introduced in April that led to the initial\u00a0H20\u00a0ban.<\/p>\n

Nvidia\u00a0is set to deliver small batches of RTX6000D to Chinese clients in September, said one of the people.<\/p>\n","protected":false},"excerpt":{"rendered":"Nvidia\u00a0is developing a\u00a0new\u00a0AI\u00a0chip\u00a0for\u00a0China\u00a0based on its latest Blackwell architecture that will be more powerful than the\u00a0H20\u00a0model it is currently…\n","protected":false},"author":2,"featured_media":8973,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[261],"tags":[4129,291,289,290,860,207,356,18,5218,19,17,778,82],"class_list":{"0":"post-8972","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-artificial-intelligence","8":"tag-advanced-micro-devices-inc","9":"tag-ai","10":"tag-artificial-intelligence","11":"tag-artificialintelligence","12":"tag-breaking-news-technology","13":"tag-business-news","14":"tag-donald-trump","15":"tag-eire","16":"tag-huawei","17":"tag-ie","18":"tag-ireland","19":"tag-nvidia-corp","20":"tag-technology"},"share_on_mastodon":{"url":"","error":""},"_links":{"self":[{"href":"https:\/\/www.europesays.com\/ie\/wp-json\/wp\/v2\/posts\/8972","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.europesays.com\/ie\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.europesays.com\/ie\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ie\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ie\/wp-json\/wp\/v2\/comments?post=8972"}],"version-history":[{"count":0,"href":"https:\/\/www.europesays.com\/ie\/wp-json\/wp\/v2\/posts\/8972\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ie\/wp-json\/wp\/v2\/media\/8973"}],"wp:attachment":[{"href":"https:\/\/www.europesays.com\/ie\/wp-json\/wp\/v2\/media?parent=8972"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.europesays.com\/ie\/wp-json\/wp\/v2\/categories?post=8972"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.europesays.com\/ie\/wp-json\/wp\/v2\/tags?post=8972"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}