{"id":57959,"date":"2025-08-10T23:31:24","date_gmt":"2025-08-10T23:31:24","guid":{"rendered":"https:\/\/www.newsbeep.com\/uk\/57959\/"},"modified":"2025-08-10T23:31:24","modified_gmt":"2025-08-10T23:31:24","slug":"ai-demand-leads-to-gpu-as-a-service-industry","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/uk\/57959\/","title":{"rendered":"AI Demand Leads to “GPU-as-a-Service” Industry"},"content":{"rendered":"

The surge of interest in AI is creating a massive demand for computing power. Around the world, companies are trying to keep up with the vast amount of GPUs<\/a> needed to power more and more advanced AI models<\/a>. While GPUs<\/a> are not the only option for running an AI model<\/a>, they have become the hardware of choice due to their ability to efficiently handle multiple operations simultaneously\u2014a critical feature when developing deep-learning models.<\/p>\n

But not every AI startup has the capital to invest in the huge numbers of GPUs now required to run a cutting-edge model. For some, it\u2019s a better deal to outsource it. This has led to the rise of a new business: GPU-as-a-service (GPUaaS). In recent years, companies like Hyperbolic<\/a>, Kinesis<\/a>, Runpod<\/a>, and Vast.ai<\/a> have sprouted up to remotely offer their clients the needed processing power.<\/p>\n

While tech giants like Amazon<\/a> or Microsoft<\/a> that offer cloud-computing services own their infrastructure, smaller startups<\/a> like Kinesis have created techniques to make the best out of the existing idle compute. <\/p>\n

\u201cBusinesses need compute. They need the model to be trained or their applications to be run; they don\u2019t necessarily need to own or manage servers,\u201d says Bina Khimani<\/a>, cofounder of Kinesis.<\/p>\n

Studies<\/a> have shown<\/a> that more than half of existing GPUs are not in use at any given time. Whether we\u2019re talking personal computers or colossal server farms, a lot of processing capacity is underutilized. What Kinesis does is identify idle compute\u2014both for GPUs and CPUs\u2014in servers<\/a> worldwide and compile them into a single computing source for companies to use. Kinesis partners with universities, data centers<\/a>, companies, and individuals who are willing to sell their unused computing power. Through a special software installed on their servers, Kinesis detects idle processing units, preps them, and offers them to their clients for temporary use.<\/p>\n

\u201cAt Kinesis, we have developed technology to pool together fragmented, idle compute power and repurpose it into a serverless, auto-managed computing platform,\u201d says Khimani. Kinesis customers may even be able to choose from where they want their GPUs or CPUs to come.<\/p>\n

AI Is Growing Faster Than Servers Can Keep Up<\/p>\n

GPUaaS is filling a growing gap in the AI industry. As learning models get more sophisticated, they need more power and an infrastructure that can process information faster and faster. In other words, without a sufficient number of GPUs, big AI models cannot operate, let alone improve. In October, OpenAI\u2019s CEO, Sam Altman, admitted<\/a> that the company was not releasing products as often as they had wished because they were facing \u201ca lot of limitations\u201d with their computing capacity.<\/p>\n

Also in October, Microsoft\u2019s CFO, Amy Woods, told<\/a> the company\u2019s investors in a conference call that demand for AI \u201ccontinues to be higher\u201d than their \u201cavailable capacity.\u201d<\/p>\n

The biggest advantage of GPUaaS is economic. By removing the need to purchase and maintain the physical infrastructure, it allows companies to avoid having to invest in servers and IT management, and instead put their resources toward improving their own deep-learning, large language, and large-vision models. It also lets customers pay for the exact amount of GPUs they use, saving the costs of the inevitable idle compute that would come with their own servers.<\/p>\n

Serverless startups like Kinesis also claim to be friendlier to the environment than traditional cloud-computing companies. By leveraging existing, unused processing units instead of powering additional servers, the company says it significantly reduces energy consumption. In the last five years, big tech companies like Google<\/a> and Microsoft have seen their carbon emissions soar<\/a> due to the amount of energy consumed by AI. In response, some have turned to nuclear energy<\/a> to sustainably power their servers. Kinesis and other new startups offer a third route in which no more servers need to be plugged in.<\/p>\n

\u201cIndustry leaders are deeply committed to sustainability,\u201d Khimani says. \u201cWith the focus on innovation and efficiency, they can optimize existing computing power that is already active and consuming energy, rather than continually adding more servers for every new application they run.\u201d<\/p>\n

The growing demand for machine learning<\/a> and colossal data consumption is turning GPUaaS into a very profitable tech sector. In 2023, the industry\u2019s market size was valued<\/a> at US $3.23 billion; in 2024, it grew to $4.31 billion. It\u2019s expected to rise to $49.8 billion by 2032.<\/p>\n

\u201cThe AI industry is rapidly advancing to a stage where the focus is shifting from merely building and training models to optimizing efficiency,\u201d Khimani says. \u201cCustomers are increasingly asking questions like, \u2018When training a new model, how can we do it extremely targeted and not consume an ocean of data that requires an enormous amount of compute and energy?\u2019\u201d<\/p>\n

From Your Site Articles<\/p>\n

Related Articles Around the Web<\/p>\n","protected":false},"excerpt":{"rendered":"The surge of interest in AI is creating a massive demand for computing power. Around the world, companies…\n","protected":false},"author":2,"featured_media":57960,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[21],"tags":[733,4323,31550,31551,86,56,54,55],"class_list":["post-57959","post","type-post","status-publish","format-standard","has-post-thumbnail","category-computing","tag-artificial-intelligence","tag-computing","tag-gpus","tag-servers","tag-technology","tag-uk","tag-united-kingdom","tag-unitedkingdom"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/posts\/57959","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/comments?post=57959"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/posts\/57959\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/media\/57960"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/media?parent=57959"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/categories?post=57959"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/tags?post=57959"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}