Qualcomm Challenges Nvidia with New AI Chips for Data Centers

Qualcomm is turning parts from cellphone chips into AI chips to rival Nvidia

Qualcomm is launching a pair of new AI chips in an attempt to challenge Nvidia's dominance in the market. On Monday, Qualcomm announced plans to release its new AI200 chip next year, followed by the AI250 in 2027 -- both of which are built on the company's mobile neural processing technology. The new chips are built for deploying AI models, rather than training them. The launch marks a notable change for Qualcomm, which has primarily made processors for mobile phones, laptops, tablets, and telecommunications equipment. As reported by CNBC, Qualcomm's AI-focused processors are based on the company's Hexagon neural processing units that power AI features in its chips for mobile devices and laptops. CNBC reports that Qualcomm's AI processors can work inside a rack with up to 72 chips functioning as a single computer, similar to Nvidia and AMD's GPUs. Qualcomm's A1200 chip features 768GB of RAM, along with performance optimized for AI inference. Meanwhile, the A1250 will come with "a generational leap in efficiency" that the company claims will allow for "much lower power consumption." Humain, the AI company that exists under Saudi Arabia's Public Investment Fund, has already announced plans to use the AI200 and AI250 to power computing systems, building on a partnership to develop AI datacenters across Saudi Arabia.

Qualcomm unveils AI200 and AI250 AI inference accelerators -- Hexagon takes on AMD and Nvidia in the booming data center realm

Qualcomm on Monday formally announced two upcoming AI inference accelerators -- the AI200 and AI250 -- that will hit the market in 2026 and 2027. The new accelerators are said to compete against rack-scale solutions from AMD and Nvidia with improved efficiency and lower operational costs when running large-scale generative AI workloads. The announcement also reaffirms Qualcomm's plan to release updated products on a yearly cadence. Both Qualcomm AI200 and AI250 accelerators are based on Qualcomm Hexagon neural processing units (NPUs) customized for data center AI workloads. The company has been gradually improving its Hexagon NPUs in the recent years, so the latest versions of these processors already feature scalar, vector, and tensor accelerators (in a 12+8+1 configuration), support such data formats as INT2, INT4, INT8, INT16, FP8, FP16, micro-tile inferencing to reduce memory traffic, 64-bit memory addressing, virtualization, and Gen AI model encryption for extra security. Scaling Hexagon for data center workloads is a natural choice for Qualcomm, though it remains to be seen what performance targets the company will set for its AI200 and AI250 units. Qualcomm's AI200 rack-scale solutions will be the company's first data-center-grade inference system powered by AI200 accelerators with 768 GB of LPDDR memory onboard (which is a lot of memory for an inference accelerator) that will use PCIe interconnects for scale-up and Ethernet for scale-out scalability. The system will use direct liquid cooling and a power envelope of 160 kW per rack, which is also an unprecedented power consumption for inference solutions. In addition, the system will support confidential computing for enterprise deployments. The solution will be available in 2026. The AI250, launching a year later, keeps this structure but adds a near-memory compute architecture to boost effective memory bandwidth by over 10 times. In addition, the system will support disaggregated inference capability that enables compute and memory resources to be dynamically shared across cards. Qualcomm positions it as a more efficient, high-bandwidth solution optimized for large transformer models, while preserving the same thermal, cooling, security, and scalability characteristics as the AI200. "With Qualcomm AI200 and AI250, we're redefining what's possible for rack-scale AI inference," said Durga Malladi, SVP & GM, Technology Planning, Edge Solutions & Data Center, Qualcomm Technologies. "These innovative new AI infrastructure solutions empower customers to deploy generative AI at unprecedented TCO, while maintaining the flexibility and security modern data centers demand." In addition to building hardware platforms, Qualcomm is also building a hyperscaler-grade, end-to-end software platform optimized for large-scale inference. The platform is set to support major ML and generative AI toolsets -- including PyTorch, ONNX, vLLM, LangChain, and CrewAI to while enabling seamless model deployment. The software stack will support disaggregated serving, confidential computing, and one-click onboarding of pre-trained models to simplify deployment. "Our rich software stack and open ecosystem support make it easier than ever for developers and enterprises to integrate, manage, and scale already trained AI models on our optimized AI inference solutions," said Malladi. "With seamless compatibility for leading AI frameworks and one-click model deployment, Qualcomm AI200 and AI250 are designed for frictionless adoption and rapid innovation." One crucial aspect about its AI200 and AI250 rack-scale solutions for inference that Qualcomm did not disclose is which processors these machines will run. The company formally began development of its own data center-grade CPUs earlier this year. While some CPU microarchitecture groundwork has probably been done by the Nuvia team before that, it is still going to take about a year to define and develop logical design, then at least six months to implement the design and tape it out, then months to bring the chip up and sample it. In short, it is reasonable to expect Qualcomm's in-house CPUs to emerge in late 2027, but rather in 2028. That said, at least the AI200 is poised to use an off-the-shelf Arm or x86 CPU, so the question is, which one?

Qualcomm announces AI accelerators and racks they'll run in

House of the Snapdragon promises - without much detail - this kit will enable coolly efficient inferencing Qualcomm has announced some details of its tilt at the AI datacenter market by revealing a pair of accelerators and rack scale systems to house them, all focused on inferencing workloads. The company offered scant technical details about its new AI200 and AI250 "chip-based accelerator cards", saying only that the AI 200 supports 768 GB of LPDDR memory per card, and the AI250 will offer "innovative memory architecture based on near-memory computing" and represent "a generational leap in efficiency and performance for AI inference workloads by delivering greater than 10x higher effective memory bandwidth and much lower power consumption." Qualcomm will ship the cards in pre-configured racks that will use "direct liquid cooling for thermal efficiency, PCIe for scale up, Ethernet for scale out, confidential computing for secure AI workloads, and a rack-level power consumption of 160 kW." In May, Qualcomm CEO Cristiano Amon offered somewhat cryptic statements that the company would only enter the AI datacenter market with "something unique and disruptive" and would use its expertise building CPUs to "think about clusters of inference that is about high performance at very low power." However, the house of the Snapdragon's announcement makes no mention of CPUs. It does say its accelerators build on Qualcomm's "NPU technology leadership" - surely a nod to the Hexagon-branded neural processing units it builds into processors for laptops and mobile devices. Qualcomm's most recent Hexagon NPU, which it baked into the Snapdragon 8 Elite SoC, includes 12 scalar accelerators and eight vector accelerators, and supports INT2, INT4, INT8, INT16, FP8, FP16 precisions. Perhaps the most eloquent clue in Qualcomm's announcement is that its new AI products "offer rack-scale performance and superior memory capacity for fast generative AI inference at high performance per dollar per watt" and has "low total cost of ownership." That verbiage addresses three pain points for AI operators. One is the cost of energy to power AI applications. Another is that high energy consumption produces a lot of heat, meaning datacenters need more cooling infrastructure - which also consumes energy and impacts cost. The third is the quantity of memory available to accelerators, a factor that determines what models they can run - or how many models can run in a single accelerator. The 768 GB of memory Qualcomm says it's packed into the AI 200 is comfortably mode than Nvidia or AMD offer in their flagship accelerators. Qualcomm therefore appears to be suggesting its AI products can do more inferencing with fewer resources, a combination that will appeal to plenty of operators as (or if) adoption of AI workloads expands. The house of Snapdragon also announced a customer for its new kit, namely Saudi AI outfit Humain, which "is targeting 200 megawatts starting in 2026 of Qualcomm AI200 and AI250 rack solutions to deliver high-performance AI inference services in the Kingdom of Saudi Arabia and globally." But Qualcomm says it expects the AI250 won't be available until 2027. Humain's announcement, like the rest of this news, is therefore hard to assess because it omits important details about exactly what Qualcomm has created and if it will be truly competitive with other accelerators. Also absent from Qualcomm's announcement is whether major hyperscalers have expressed any interest in its kit, or if it will be viable to run on-prem. The announcement does, however, mark Qualcomm's return to the datacenter after past forays focused on CPUs flopped. Investors clearly like this new move as the company's share price popped 11 percent on Monday. ®

Qualcomm shares jump as it launches new AI chip to rival Nvidia

Qualcomm's shares surged as much as 20 per cent on Monday as the US chip company launched its first data-centre processors for artificial intelligence, seeking to grab a share of a multibillion-dollar market dominated by Nvidia. Saudi Arabia's Humain, an AI company backed by the kingdom's Public Investment Fund, will be Qualcomm's first customer, as they build on a partnership announced in May. Qualcomm's stock, which was trading around 12 per cent higher by midday in New York, made its biggest one-day move for several years in response to the news, adding tens of billions of dollars to push its market capitalisation above $200bn. Investors see the chip giant, which is best known for its smartphone processors, moving to diversify its products and benefit from the AI infrastructure boom, including a recent surge in investment in "sovereign AI". So far, Nvidia has been the primary beneficiary of the AI boom, with analysts estimating that it holds more than three quarters of the market for the specialised processors needed to train and run large language models, the systems behind OpenAI's ChatGPT and Google's Gemini. Qualcomm's launch is the latest example of how rivals are trying to narrow that lead, after OpenAI struck AI chip deals with AMD and Broadcom in recent weeks. Humain plans to deploy 200 megawatts of Qualcomm's new AI accelerators starting in 2026, as the Gulf state seeks to position itself as a hub for artificial intelligence. "By establishing advanced AI data centres powered by Qualcomm's industry-leading inference solutions, we are helping the kingdom create a technology ecosystem that will accelerate its AI ambitions of becoming a hub of intelligent computing," said Qualcomm chief executive Cristiano Amon. The announcement follows a state visit to the Middle East by US President Donald Trump in May, when he was accompanied by US tech leaders including Amon. Qualcomm and Humain at the time struck a memorandum of understanding to deliver advanced AI data centres. Qualcomm's AI200 and AI250 chips will launch in 2026 and 2027 respectively, promising to help AI applications run faster. The products will be available in rack-scale, liquid-cooled formats, mirroring how Nvidia and its competitors have moved towards offering multiple chips connected up inside a server rack. Qualcomm says it is now committed to annual launches of AI chips, following a pattern established by Nvidia. The AI250 will also offer a new memory architecture that Qualcomm says will bring a "generational leap in efficiency and performance". Memory has emerged as one of the key constraints on the speed and capability of AI chips. Massive investment in data centres by Big Tech groups as well as nation states has catapulted Nvidia past a record $4tn valuation. But rivals such as AMD, as well as Nvidia's own customers including Amazon and OpenAI, are developing their own AI processors. Shares in UK-based Arm, whose chip blueprints are used by Qualcomm to design its chips, also rose around 4.5 per cent on Monday. Nvidia shares were up around 2.5 per cent.

Qualcomm announces AI chips to compete with AMD and Nvidia

The AI chips are a shift from Qualcomm, which has thus far focused on semiconductors for wireless connectivity and mobile devices, not massive data centers. Qualcomm said that both the AI200, which will go on sale in 2026, and the AI250, planned for 2027, can come in a system that fills up a full, liquid-cooled server rack. Qualcomm is matching Nvidia and AMD, which offer their graphics processing units, or GPUs, in full-rack systems that allow as many as 72 chips to act as one computer. AI labs need that computing power to run the most advanced models. Qualcomm's data center chips are based on the AI parts in Qualcomm's smartphone chips called Hexagon neural processing units, or NPUs. "We first wanted to prove ourselves in other domains, and once we built our strength over there, it was pretty easy for us to go up a notch into the data center level," Durga Malladi, Qualcomm's general manager for data center and edge, said on a call with reporters last week. The entry of Qualcomm into the data center world marks new competition in the fastest-growing market in technology: equipment for new AI-focused server farms. Nearly $6.7 trillion in capital expenditures will be spent on data centers through 2030, with the majority going to systems based around AI chips, according to a McKinsey estimate.

Qualcomm steps into the AI infrastructure race with new AI200 and AI250 accelerators

Serving tech enthusiasts for over 25 years. TechSpot means tech analysis and advice you can trust. Looking ahead: Qualcomm is positioning itself to capture a share of the next decade's data center spending surge by combining its expertise in mobile efficiency with a scalable rack design. The company is no longer presenting itself solely as a smartphone chipmaker; it is now entering the infrastructure race that will determine which hardware powers the next generation of AI applications. Qualcomm is expanding beyond its roots in mobile technology, directly challenging the companies dominating artificial intelligence hardware. The semiconductor firm has announced that it will enter the high-end data center market with two new AI accelerator chips, marking its most ambitious move yet into the computing infrastructure that underpins the current AI boom. The company plans to release two products: the AI200 and AI250, both designed for the inference side of AI deployment. The AI200 will be available commercially in 2026, followed by the AI250 in 2027. Both chips can be configured as full, liquid-cooled server racks for large-scale data centers. Qualcomm also committed to updating its AI data center hardware annually. While Qualcomm is best known for its mobile-oriented Snapdragon processors and wireless connectivity chips, the new line of AI accelerators borrows technology from its Hexagon neural processing units, which are optimized for low-power machine learning tasks in smartphones. Company executives said these same design efficiencies could make Qualcomm's rack-scale systems cost-competitive with data center offerings from Nvidia and AMD, whose GPUs currently dominate the field. "The idea was to first prove ourselves in mobile and edge computing before scaling up to the data center level," Durga Malladi, general manager for Qualcomm's data center and edge business, said in an earnings call. "Our architecture allows customers to either buy our complete rack system or combine our chips with their own designs." The hardware structure of Qualcomm's new systems resembles Nvidia's HGX and AMD's Instinct-based platforms: large racks holding dozens of interconnected accelerators that function as a single compute unit. Each rack consumes approximately 160 kilowatts of power, comparable to current high-performance GPU clusters. The company emphasized what it calls superior performance per dollar per watt. Unlike Nvidia's H100 GPUs, which specialize in both training and inference, Qualcomm is targeting inference workloads only. This includes generating text using pretrained models or supporting interactive applications that require real-time processing. Malladi said these tasks represent an increasing share of AI data center usage. The company declined to disclose pricing or the number of NPUs a single rack could hold but confirmed that its AI cards support 768 gigabytes of memory, an amount exceeding the capacity of comparable products from Nvidia and AMD. Qualcomm said it has developed a new memory management architecture designed to improve speed and reduce energy consumption during inference operations. Qualcomm's entry signals an intensifying effort to expand the AI semiconductor ecosystem, potentially loosening Nvidia's dominance in the sector. The company already has its first major client: Saudi Arabia-based Humain, which will deploy Qualcomm's AI200 and AI250 systems starting in 2026. The partnership could power data center capacity equivalent to 200 megawatts once fully operational. Qualcomm also said its systems will be available as discrete components for hyperscale cloud operators that prefer to design and assemble their own racks. Its CPUs and accelerator cards could, in principle, be supplied even to rivals. "Our goal is to give customers choices - take the full rack or mix and match," Malladi said.

Qualcomm's new AI systems promise 10x bandwidth, lower power use

"These innovative new AI infrastructure solutions empower customers to deploy generative AI at unprecedented TCO, while maintaining the flexibility and security modern data centers demand." Malladi added that Qualcomm's rich software stack and open ecosystem support will make it easier for developers and enterprises to integrate, manage, and scale already trained AI models. The platform supports leading AI frameworks and one-click model deployment, enabling "frictionless adoption and rapid innovation." The company's hyperscaler-grade AI software stack underpins the hardware, offering end-to-end support from application to system software layers. It is optimized for inference across major machine learning frameworks, generative AI platforms, and inference engines. Developers will be able to seamlessly onboard Hugging Face models via Qualcomm's Efficient Transformers Library and AI Inference Suite: tools designed for operationalizing AI through ready-to-use applications, agents, and APIs. Qualcomm expects the AI200 to become commercially available in 2026, followed by the AI250 in 2027. The company said it plans to maintain an annual cadence of data center product updates, focusing on performance, energy efficiency, and continuous innovation in AI inference.

Qualcomm aims for a spot in the AI data center with new chips

Why it matters: Unprecedented demand has created a fresh opportunity for Nvidia's rivals -- if they can deliver sufficiently powerful chips. Driving the news: Qualcomm announced two chips: the AI200, due out next year; and a follow-up chip -- AI250 -- due the following year, with what Qualcomm promises is a ground-breaking new memory architecture. * "It improves the memory bandwidth by a factor of more than 10," Qualcomm senior vice president Durga Malladi said in a briefing with reporters. * While Qualcomm typically delivers only chips to customers, here it has also designed a rack-level system designed to plug directly into a data center. Of note: Saudi AI company Humain will be the first customer for the new chips, with plans to bring on 200 megawatts of compute starting in 2026. Between the lines: Qualcomm is talking up the energy efficiency of its chips, saying that they will be cheaper over the long term. * Earlier this year, the company touted a study showing AI compute on devices running its mobile chips dramatically decreased power consumption. Flashback: It's not Qualcomm's first effort to crack into the data center.

Qualcomm to take on Nvidia with its own AI chips

On October 27, Qualcomm released a new series of artificial intelligence chips to compete with the market leader Nvidia, as the race to cash in on the massive AI datacenter buildout heats up. If successful, Qualcomm, a San Diego-based tech giant, could win a spot in the data centers powering AI as customers look for alternatives to Nvidia, which controls close to 90% of the AI chip market. The Qualcomm AI200 is scheduled to be the first chip in the series that will be commercially available in 2026, which will be followed by the AI250 chip in 2027. The company's stock surged 20% after news of entering the data center market. It plans to sell its purpose-built AI server racks containing dozens of AI chips that can be installed in data centers, and will also sell just the standalone AI chips which enterprises can buy and plug into their existing servers. "Our rich software stack and open ecosystem support make it easier than ever for developers and enterprises to integrate, manage, and scale already trained AI models on our optimized AI inference solutions," said Durga Malladi, a senior vice president at Qualcomm. Qualcomm is best-known for its chips used to power smartphones. It is one of the latest entrants into the AI chipmaking arena, joining Intel and AMD, to compete with Nvidia. Qualcomm is positioning itself as an energy efficient chip that will cost much less to run in the long term. These companies see an opportunity in building inference chips, which are used when a trained AI model runs real-time computations to generate outputs such as answer questions or generating images. Demand for such AI inference chips has increased with wider adoption and newer use cases, with companies such as Amazon, Google, and Microsoft creating their own AI chips. Nearly $7 trillion of capital expenditure will be spent on datacenters through 2030, according to a Mckinsey estimate. "It makes sense that Qualcomm wants to diversify beyond smartphones and get into this game," says Austin Lyons, analyst and founder of Chipstrat, a semiconductor publication. "This is a good and different vector -- not just consumer products, but datacenter." OpenAI, in September, signed a $10-billion deal with Broadcom to design custom AI chips, and has invested in AMD with a commitment to buy its MI450 AI chips. Qualcomm also signed Saudi Arabia's Humain, an AI company backed by the country's sovereign wealth fund, as its first customer for the new series of chips. The chips will be deployed in datacenters of Humain in 2026. Humain plans to launch a $10-billion venture fund, and in May, picked another California chipmaker, Groq, to supply inference chips for its datacenters. Abu Dhabi-backed G42, an AI development holding company -- that holds a stake in U.S. chipmaker Cerebras Systems -- will build the 5 gigawatt UAE-US AI campus that was announced during President Trump's visit in May. Gulf countries have emerged as a power player in AI, as Trump's White House overhauled Biden-era chip export restrictions and brokered multi-billion dollar deals for the U.S. to supply advanced chips required to fuel its AI ambitions. 2025 Los Angeles Times. Distributed by Tribune Content Agency, LLC.

Qualcomm debuts AI200, AI250 data center AI chips - SiliconANGLE

Qualcomm Technologies Inc. today introduced a pair of artificial intelligence chips designed for use in data centers. Shares of the company jumped more than 15% on the news at one point. Its stock is currently up about 11%. The two new chips, the AI200 and AI250, will be sold as part of accelerator cards that can be plugged into servers. Qualcomm says the processors are geared towards inference use cases that prioritize cost-efficiency. According to the company, the chips will provide that cost-efficiency in the form of "high performance per dollar per watt." Qualcomm didn't disclose how much power the accelerators will use or what processing speeds they are expected to provide. However, the company did specify that they're based on its Hexagon architecture. The architecture underpins the neural processing units that Qualcomm ships with its consumer systems-on-chip. A Hexagon-based NPU can be found in, among other products, the company's flagship Snapdragon 8 Elite Gen 5 smartphone processor. The chip is made using a 3-nanometer manufacturing process. Its NPU includes 20 cores based on three different designs that can process up to 220 tokens per second. Qualcomm has also integrated Hexagon into its connected device chips. Last year, the company introduced an NPU-equipped processor designed to power Wi-Fi routers. The built-in NPU can run local AI models optimized for tasks such as filtering malicious network traffic and optimizing routers' power usage. Qualcomm's new AI200 and AI250 chips likely have a significantly higher core count than its consumer NPUs. Nvidia Corp.'s Blackwell Ultra includes over 20,000 cores, or about 1,000 times more than the NPU in the Snapdragon 8 Elite Gen 5. The AI200, the less advanced of Qualcomm's AI chips, offers 768 gigabytes of LPDDR memory. LPDDR is a type of RAM used mainly in mobile devices. It consumes less power than the DDR5 memory found in servers and provides lower memory bandwidth. Memory bandwidth, which determines how fast data moves between a chip's cores and the attached RAM, heavily influences AI models' inference speeds. Qualcomm says that the AI250 will offer more than 10 times the memory bandwidth of the AI200. It's possible the chipmaker plans to achieve that speedup by swapping the AI200's relatively slow LPDDR memory with a more performant RAM variety. One candidate is HBM memory, which is widely used in data center AI processors. Qualcomm disclosed that the AI250 and AI200 will include a confidential computing feature. The technology, which is also supported by Nvidia's Blackwell Ultra, splits the memory of an AI chip into multiple encrypted sections. Only the application that uses a given memory section can read its contents. Qualcomm plans to ship its AI chips as part of water-cooled compute racks. The appliances use PCIe to link together their internal components and Ethernet to provide connectivity between systems. It's possible the racks will include the server-grade central processing units that Qualcomm is currently developing. Alongside machine learning accelerators, AI appliances include CPUs that manage general-purpose computing tasks such as running the operating system. Nvidia already ships its rack-scale DGX AI appliances with internally-developed CPUs. Qualcomm will launch the AI200 and AI250 in 2026 and 2027, respectively. The company plans to refresh its data center AI processor lineup annually going forward.

Qualcomm (QCOM) stock soars as company challenges Nvidia with new AI data center chips

Shares of Qualcomm (NASDAQ: QCOM) surged on Monday after the company announced a bold new strategy to enter the AI data center market, unveiling a series of chips designed to rival Nvidia and AMD for certain AI-related tasks. The news sent the Qualcomm stock price soaring, signaling a new chapter for the semiconductor giant as it pivots toward the high-stakes world of enterprise AI. This strategic move is a clear response to the unprecedented demand for AI computing power, creating a fresh opportunity for new players to challenge the market leaders. For investors watching QCOM stock, this represents a significant push beyond the company's traditional mobile chip business. On Monday, Qualcomm announced two new accelerator chips and a full rack-level system designed to plug directly into a data center. Both solutions are designed for generative AI inference workloads and come with comprehensive software support, including for popular AI frameworks and one-click deployment of Hugging Face models. The company is positioning these products as cost-effective alternatives for data centers running AI, with a focus on energy efficiency and a lower total cost of ownership. Saudi AI company Humain will be the first customer for the new chips, with plans to bring 200 megawatts of compute online starting in 2026. This announcement marks Qualcomm's second major attempt to break into the data center market. Its first effort with the Centriq family of processors in 2017 failed to make a significant dent in Intel's market share, and the company exited the market a year later. This time, however, the market is different. The explosive growth of AI has created a seemingly insatiable demand for powerful, efficient chips, a market currently dominated by Nvidia. The market's reaction was overwhelmingly positive. The Qualcomm stock price jumped 12% on Monday following the announcement, reflecting investor confidence in the company's new direction. This surge pushed QCOM shares to a new 52-week high of $182.23, giving the company a market capitalization of over $185 billion. With this new initiative, Qualcomm is making a direct play for a piece of the lucrative AI data center pie. While commercial availability of the new chips is still a year away (the AI200 in 2026 and the AI250 in 2027), the company has committed to an annual release cycle for its data center AI roadmap. The move represents a significant strategic expansion for Qualcomm and a new competitive threat to Nvidia's dominance. For investors in QCOM stock, it signals a clear commitment to capitalizing on the biggest trend in technology today: Artificial intelligence.

Qualcomm's AI200 turns up the heat on Nvidia -- and puts inference economics in the spotlight - SiliconANGLE

Qualcomm's AI200 turns up the heat on Nvidia -- and puts inference economics in the spotlight Qualcomm Inc. shares spiked as much as 20% early today after the company unveiled new data center artificial intelligence accelerators, the AI200 and AI250, aimed squarely at Nvidia Corp.'s inference stronghold with its graphics processing units. According to today's reporting, AI200 is slated to ship in 2026, with AI250 following in 2027, and both will come as standalone components or add‑in cards that slot into existing servers. The move expands Qualcomm's strategy from AI PCs and edge devices into cloud and enterprise inference at scale -- a battleground where cost, power and software maturity decide winners. Here is my Breaking Analysis from the Cube Community in context from our reporting, interviews and research. Qualcomm is playing the right game at the right time. Inference is the AI profit center and it's increasingly heterogeneous. If AI200/AI250 deliver competitive latency, model density, and perf‑per‑watt -- with a developer‑friendly stack -- Qualcomm can carve out meaningful share in a market that wants credible alternatives to GPU‑only designs. The company's history in low‑power, Arm‑based compute, its momentum in AI PCs, and prior AI100 deployments provide a foundation. The hurdle is software gravity and ecosystem depth, where Nvidia still sets the pace. Our bottom line: 2026-2027 will see an accelerated shakeout in inference silicon. Qualcomm's announcement signals it plans to be in that final round -- and enterprises should welcome the added optionality.

Qualcomm Announces New AI Chips in Data Center Push

Qualcomm on Monday unveiled two artificial intelligence chips for data centers that will be available next year, diversifying beyond a stagnant smartphones market and sending its shares up 20 percent. The share gain following the news underscores strong enthusiasm for the company's AI bets while the smartphone chipmaker geared up to compete against Nvidia's AI data center heft. The new chips, called AI200 and AI250, are designed for improved memory capacity and running AI applications, or inference, and will be commercially available in 2026 and 2027, respectively. Global investment in AI chips has soared as cloud providers, chipmakers and enterprises rush to build infrastructure capable of supporting complex, large language models, chatbots and other generative AI tools. Qualcomm said the new chips support common AI frameworks and tools and played up cost-savings for enterprises. The company also unveiled racks based on the new chips, as Nvidia and AMD move from selling chips to providing larger data center systems. Though competition against Nvidia has been heating up, the high costs of switching chip providers and superior performance of Nvidia processors has made it difficult for new entrants to gain traction. Qualcomm said Humain, an AI startup launched by Saudi Arabia's sovereign wealth fund, will deploy 200 megawatts of its new AI racks starting in 2026. "Qualcomm's entry and major deal in Saudi Arabia prove the ecosystem is fragmenting because no single company can meet the global, decentralized need for high-efficiency AI compute," said Joe Tigay, portfolio manager of the Rational Equity Armor Fund. Qualcomm Diversifies Qualcomm is the world's largest supplier of modem chips that enable smartphones to connect to wireless data networks. But it has been diversifying its business to reduce dependence on the smartphone market, which makes up a majority of its sales after losing Huawei as a major customer and client Apple's efforts to develop in-house chips. Over the last two years, it has entered the personal computer market, competing against Intel and AMD to sell chips that power Windows-based laptops.

Qualcomm Stock Is Soaring Today After Chipmaker Makes a Big AI Move

The move could mean more competition for leading AI chipmakers such as Nvidia and AMD. Qualcomm is joining the competition to supply chips for AI data centers. Qualcomm (QCOM), which has largely focused on chips for mobile phones, laptops and other consumer devices to date, on Monday unveiled two AI accelerator chips for data centers: the AI200, which is set for release in 2026, and the AI250, which is planned for 2027. Shares of Qualcomm were up more than 13% in recent trading, pacing gainers in the S&P 500 and Nasdaq Composite. With Monday's gains, they've added about one-quarter of their value in 2025. The move marks a big step for Qualcomm, which said that it will be committed to a data center roadmap with an annual cadence for major releases or updates going forward. It could also mean more competition for leading AI chipmakers such as Nvidia (NVDA) and Advanced Micro Devices (AMD), with several analysts suggesting in recent weeks that they see Nvidia's dominant market share diminishing over time as competition and demand grows. Shares of Nvidia were up 2.5% in recent trading amid a broader market rally, while AMD shares were up slightly after hitting an all-time high at the open.

Qualcomm accelerates data center push with new AI chips launching next year

Qualcomm is expanding into the AI infrastructure market with two new data center chips, AI200 and AI250, set for commercial availability in 2026 and 2027. This move aims to diversify its business beyond smartphones and capitalize on the booming demand for AI hardware. The company also announced accelerator cards and racks based on these new chips. Qualcomm on Monday unveiled two artificial intelligence chips for data centers, with commercial availability from next year, as it pushes to diversify beyond smartphones and expand into the fast-growing AI infrastructure market. Shares of Qualcomm surged nearly 15% on the news. The new chips, called AI200 and AI250, are designed for improved memory capacity and running AI applications, or inference, and will be available in 2026 and 2027, respectively. Global investment in AI chips has soared as cloud providers, chipmakers and enterprises rush to build infrastructure capable of supporting complex, large language models, chatbots and other generative AI tools. Nvidia chips, however, underpin much of the current AI boom. Qualcomm, to strengthen its AI portfolio, agreed to buy Alphawave in June, which designs semiconductor tech for data centers, for about $2.4 billion. In May, Qualcomm also said it would make custom data center central processing units that use technology from Nvidia to connect to the firm's artificial intelligence chips. Qualcomm said the new chips support common AI frameworks and tools, with advanced software support, and added they will lower the total cost of ownership for enterprises. The San Diego-based company also unveiled accelerator cards and racks based on the new chips. Earlier this month, peer Intel announced a new artificial intelligence chip called Crescent Island for the data center that it plans to launch next year.

Qualcomm's New AI Rack-Scale Solution Actually Uses LPDDR Mobile Memory Onboard, Boldly Hoping to Take on NVIDIA and AMD

Qualcomm has announced its latest AI chips, which are designed to scale up to a purpose-built rack-level AI inference solution, but interestingly, they employ mobile memory onboard. Qualcomm has come a long way from being a mobile-focused firm, and in recent years, the San Diego chipmaker has expanded into new segments, including consumer computing and AI infrastructure. Now, the firm has announced its newest AI200 and AI250 chip solutions, which are reportedly designed for rack-scale configurations. This not only marks the entry of a new player in a segment dominated by NVIDIA and AMD, but Qualcomm has managed to find a unique implementation by utilizing mobile-focused LPDDR memory. Before we delve into the specifics of the newly announced AI chips, let's examine the use of LPDDR memory compared to the more traditional HBM solution. Qualcomm's new products offer up to 768 GB of LPDDR on the accelerator package, which is significantly higher than the industry's HBM capacity. The main reason this venture looks attractive is that it reduces data-movement energy and cost, a key advantage that the firm calls a "near-memory" approach. Here are the traditional improvements the firm gets by employing LPPDR over HBM: While this implementation sounds optimistic, Qualcomm's rack-scale solutions still fall short when compared to mainstream options from NVIDIA/AMD, simply because avoiding HBM use results in lower memory bandwidth, higher latency due to a narrow interface, and, most importantly, utilizing an immature memory stack in 24/7 high-heat server environments. However, the intention of the San Diego firm here is to provide companies with a capable inferencing option, and the use of LPDDR certainly achieves this goal, but it does limit these rack-scale configurations to a specific application. Apart from this, the AI200 and AI250 chip solutions feature direct liquid cooling, PCIe/Ethernet protocols, and a rack-level power consumption of 160 kW, which is a pretty low figure for a modern-day solution. More importantly, the chips onboard employ the firm's Hexagon NPUs, which are widely expanding in terms of inferencing capabilities, supporting advanced data formats as well as inference-focused features. Interestingly, the pivot towards bringing capable inferencing solutions to the market is being done by a lot of compute providers, with one of the more recent examples being Intel with its 'Crescent Island' solution and NVIDIA introducing a new Rubin CPX AI chip. Qualcomm apparently recognizes that the inferencing segment is gaining market spotlight, which is why the AI200 and AI250 chip solutions are a sensible approach here. However, for modern training or large workloads, these racks would probably be the last choice. It's exciting to see competition emerging in the AI space, and by the looks of it, retailers took the announcements with quite some optimism.

Qualcomm Unveils New AI Chips To Compete In Data Center Race - Qualcomm (NASDAQ:QCOM)

Qualcomm Technologies, Inc. (NASDAQ:QCOM) announced on Monday the launch of its next-generation artificial intelligence inference-optimized solutions for data centers, namely the Qualcomm AI200 and AI250 chip-based accelerator cards and racks. QCOM is showing upward movement. Get the complete analysis here Building on the company's leadership in Neural Processing Unit (NPU) technology, these solutions offer rack-scale performance and superior memory capacity for fast generative AI inference, delivering high performance per dollar per watt, Qualcomm said. Qualcomm AI200 introduces a purpose-built rack-level AI inference solution designed to deliver low total cost of ownership (TCO) and optimized performance for large language & multimodal model (LLM, LMM) inference, as well as other AI workloads. Also Read: Qualcomm And Valeo Broaden Collaboration To Speed Hands Off Driving Features Performance It supports 768 GB of LPDDR per card, offering higher memory capacity and lower cost, while enabling exceptional scale and flexibility for AI inference. The Qualcomm AI250 solution will debut with an innovative memory architecture based on near-memory computing, providing a generational leap in efficiency and performance for AI inference workloads by delivering more than 10 times higher effective memory bandwidth and significantly lower power consumption. This enables disaggregated AI inferencing for efficient utilization of hardware while meeting customer performance and cost requirements. Both rack solutions feature direct liquid cooling for thermal efficiency, PCIe for scale up, Ethernet for scale out, confidential computing for secure AI workloads, and a rack-level power consumption of 160 kW. Qualcomm AI200 and AI250 will be commercially available by 2026 and 2027, respectively. Competition Qualcomm's AI accelerator rivals include Nvidia Corp's (NASDAQ:NVDA) H100 and H200 chips, Advanced Micro Devices, Inc's (NASDAQ:AMD) Instinct MI300X accelerators, and Intel Corp's (NASDAQ:INTC) Gaudi accelerators. Alphabet Inc. (NASDAQ:GOOGL) Google has developed its own Tensor Processing Units (TPUs), which are optimized for popular machine learning frameworks, including TensorFlow and PyTorch. Amazon.com Inc. (NASDAQ:AMZN) Amazon Web Services (AWS) created Inferentia chips to help customers scale machine learning applications more effectively. Price Action: Qualcomm stock is trading higher by 0.97% to $170.58 premarket at last check Monday. Read Next: Qualcomm And Google Cloud Forge AI Alliance To Transform Cars Into Smart Agents Photo via Qualcomm QCOMQualcomm Inc$169.590.38%OverviewAMDAdvanced Micro Devices Inc$253.680.30%AMZNAmazon.com Inc$227.141.31%GOOGLAlphabet Inc$265.582.18%INTCIntel Corp$39.904.22%NVDANVIDIA Corp$190.892.48%Market News and Data brought to you by Benzinga APIs

Does Qualcomm's Entry Into the AI Chip Race Spell Trouble for Nvidia? | The Motley Fool

The maker of affordable high-performance mobile processors has made an unlikely leap into territory largely controlled by a much more prolific competitor. Long-standing lines that have distinguished chipmakers from one another are being crossed. Namely, Qualcomm (QCOM +2.16%) -- largely focused on computing processors for mobile devices -- is entering the artificial intelligence arena that Nvidia (NVDA 0.20%) dominates. The company said as much on Monday, unveiling two new processors purpose-built for AI data centers. The question is, will Qualcomm's unlikely foray into the business prove disruptive to Nvidia, and, for that matter, relative newcomer Advanced Micro Devices (AMD +0.50%)? Maybe. But, first things first. OK, it's not exactly a jaw-dropping shocker. The $2.4 billion acquisition of AI inferencing specialist Alphawave Semi, announced in June, was explicitly touted as a deal that "provides key assets for Qualcomm's expansion into data centers." And, considering Global Market Insights' forecast of 15% annual growth in the worldwide data center chip market from around $16 billion now to more than $60 billion by 2034, there's just too much money on the table to pass up. Nevertheless, seeing confirmation of a specific product makes it very real for existing and would-be shareholders. That's why Qualcomm stock jumped more than 11% on Monday in response to the news. And it's encouraging news to be sure. The AI200 chip-based accelerator cards (and racks) expected to debut next year will offer "superior memory capacity for fast generative AI inference at high performance per dollar per watt." The AI250 slated for launch the following year will do the same, but even better, putting Qualcomm squarely in a space other than the high-performance mobile processor market, where it's been focused for years now. Simply getting into a business, of course, doesn't necessarily mean the industry is interested in abandoning more proven providers and purchasing your technology instead. Do Nvidia and its shareholders (and to a lesser degree, AMD and its investors) have something to worry about here? There's no outright confirmed figure of Nvidia's share of the artificial intelligence accelerator market. Given its early entry in the race with purpose-built processors, however, no one seems to dispute estimates that the number could be as high as 90%. This sort of commanding lead can't last forever, though. A competitor largely just needs to step up its design and marketing efforts to make a dent in Nvidia's dominance. That's what AMD finally did in earnest late last year, unveiling its MI325X chip meant to compete with Nvidia's Blackwell processors, which -- at the time anyway -- were its AI data center workhorse. And it's done reasonably well with this lineup. AMD's second-quarter data center revenue of $3.2 billion was up 14% year over year despite headwinds in China, proving Nvidia's grip on the market isn't exactly ironclad. That's not the only proof that Nvidia can be beaten on the artificial intelligence data center front either. Data center owners/operators like Amazon, Alphabet's Google, and Microsoft are also increasingly bypassing Nvidia and instead opting to work directly with chip developers like Broadcom (AVGO 1.71%) and Marvell Technology (MRVL +5.84%) to manufacture their own custom silicon. Google's Tensor processing units, serving as the digital brain for several of its training and inference platforms, were actually co-developed with Broadcom, for instance. Notably, artificial intelligence newcomer Anthropic is a key user of Google's cloud-provided Tensor technology. Kevin Scott, Microsoft's chief technology officer, has again stated that his company aims to increase its use of proprietary AI chips, thus decreasing its dependence on Nvidia's commercial offerings. None of this dynamic is of any tangible benefit to Qualcomm, just as none of it poses a direct or immediate threat to Nvidia. Indirectly, however, at the very least, it confirms that Nvidia's leadership of the AI semiconductor market is fading. There are finally alternatives out there that key players in the industry are choosing. Qualcomm's entry into the race only adds to the mix of choices that chip away at Nvidia's dominance. The challenge for investors here is largely just one of timing and relativity. Clearly, competitors are coming to the market, but it could take years for the entire AI data center industry to wean itself from Nvidia's wares, which it's become very familiar with. And the artificial intelligence hardware market is also still growing like crazy in the meantime. Even if it's winning less business in the future, Nvidia could still win enough of this growth to continue pumping up its top and bottom lines. Ditto for AMD. Read between the lines, though, through a more nuanced lens. Like weight-loss drugs, e-commerce, solar panels, electric vehicles, and a slew of other industries, the capitalism-driven marketplace isn't going to let a single powerhouse dominate a lucrative business like this one forever. It's a prospective problem for Nvidia shareholders largely because much of the stock's premium pricing of late has been rooted in its dominance of the AI processor market. Now that reason is starting to crumble, even if only a little for now. Sheer uncertainty can take a surprisingly sizable toll on any stock's value. That being said, while Qualcomm's entry into the artificial intelligence chip business is yet another argument against sticking with a stake in Nvidia, in and of itself, it isn't a reason to step into a position in Qualcomm. Although it has one customer lined up -- Saudi Arabia's AI company Humain -- other players may not be in any particular hurry to test-drive its fairly new tech. The company's foray into this market is simply going to further democratize the AI chip industry.

Qualcomm stock skyrockets 19%, hits 52-week high after it announces chips to take on Nvidia and AMD

Qualcomm stock just soared 19% -- its biggest jump in over a year. The reason? Qualcomm announced new AI accelerator chips, the AI200 and AI250, built to rival Nvidia and AMD in powering AI data centers. With a $182 billion market cap and entry into a $6.7 trillion AI infrastructure boom, Qualcomm is no longer just a smartphone chipmaker -- it's going all-in on energy-efficient AI hardware that could redefine the next era of AI computing. Analysts broadly see Qualcomm's stock as a moderate buy with significant upside potential.

Qualcomm looks to take on Nvidia, AMD as it enters AI accelerator market (QCOM:NASDAQ)

Qualcomm (NASDAQ:QCOM) announced its entry into the artificial intelligence accelerator market on Monday, as it looks to take on industry heavyweights Nvidia (NASDAQ:NVDA) and AMD (NASDAQ:AMD). The Cristiano Amon-led company unveiled its Qualcomm A1200 and AI250 chip-based accelerator cards and racks, all Qualcomm's entry with the AI200 and AI250 aims to compete with industry leaders, leveraging high memory capacity, lower costs, and innovative architecture; however, it is starting from a much smaller data center revenue base than Nvidia and AMD. Qualcomm targets the sizeable AI accelerator market, which AMD's CEO recently estimated could exceed $500 billion. Qualcomm is diversifying beyond handsets into automotive, IoT, and now AI accelerators; recent quarters showed revenue growth in automotive and IoT, indicating progress broadening its revenue sources.

Qualcomm Enters AI Chip Market as Rival to Nvidia and AMD | PYMNTS.com

The San Diego-based company said the AI200 will be available in 2026 and the AI250 in early 2027. The processors are built for the "inference" phase of AI, where trained models are put to work on real-world tasks rather than being developed. Qualcomm said the chips can be installed individually or in full data-center racks and will support popular AI software frameworks to simplify deployment for businesses. Inference already represents a growing share of total computing demand and is expected to overtake training by 2026 as companies embed AI into customer support, financial forecasting and logistics workflows. Qualcomm said its new chips are optimized for performance per watt, a measure of how efficiently they process AI tasks. Internal testing cited by CNBC showed that an AI200 rack can deliver equivalent output using up to 35% less power than comparable GPU-based systems savings that could lower annual energy costs by millions of dollars for large data-center operators. Competitors are also expanding their offerings. AMD's MI325X accelerator, launched in September, is built for high-memory AI workloads, while Intel's Gaudi 3 emphasizes open-source integration. Qualcomm's approach differs by offering rack-scale inference systems, allowing enterprises to install complete configurations rather than assembling components. The company also announced a partnership with Saudi-based startup Humain, which plans to deploy about 200 megawatts of Qualcomm-powered AI systems starting in 2026, according to Investors.com. Qualcomm said the collaboration demonstrates the chips' readiness for enterprise-scale workloads across sectors including finance, manufacturing and healthcare. Qualcomm's move into AI infrastructure reflects its strategy to diversify beyond smartphones -- a market that has matured in recent years. The company completed a $2.4 billion acquisition of U.K.-based Alphawave IP Group in June to expand its connectivity and systems integration capabilities for large computing installations, Reuters reported. The launch positions Qualcomm in direct competition with Nvidia and Advanced Micro Devices (AMD), which dominate AI data-center hardware. As The Wall Street Journal noted, Qualcomm's entry signals that chipmakers are racing to capture enterprise demand as more companies build their own AI infrastructure rather than relying entirely on cloud providers. Qualcomm President Cristiano Amon told CNBC that the company aims to make AI "cost-efficient at scale," drawing on its experience building power-efficient mobile chips to improve energy performance in large computing environments. "The next stage of AI will be about running it everywhere efficiently," Amon said. Running AI systems at scale is costly. Every time a generative model answers a question, analyzes data or processes a transaction, it consumes computing power and electricity. Qualcomm said its new chips are engineered to deliver high performance with lower power use, potentially helping businesses manage AI expenses more predictably. While Nvidia continues to dominate AI training, Qualcomm's strategy targets inference, the layer where models perform the work. Nvidia continues to dominate training chips, but its near monopoly on inference is already eroding as firms like AMD, Intel and now Qualcomm introduce alternatives built around energy efficiency and modular deployment. For enterprises, the arrival of new chip suppliers could translate into more options for sourcing infrastructure and lower barriers to scaling AI tools. The data-center market is also expanding rapidly. Qualcomm's focus on power efficiency and cost predictability aims to attract enterprise buyers who measure success by operational stability and long-term total cost of ownership, rather than peak computing speed. If these new entrants succeed, enterprises could benefit from greater supply resilience and more competitive pricing in the years ahead. A more diverse chip supply chain may ease the GPU shortages that have constrained enterprise AI expansion, while competition among hardware vendors could lower infrastructure costs across the industry. As PYMNTS has reported, the global spending on AI infrastructure could exceed $2.8 trillion through 2029.

Why Qualcomm Stock Jumped Today | The Motley Fool

Shares of Qualcomm (QCOM +11.08%) popped on Monday after the tech giant said it was entering the booming artificial intelligence (AI) chip market. As of 3:15 p.m., Qualcomm's stock price was up more than 12%, after rising as much as 22% earlier in the day. The semiconductor maker debuted its AI200 and AI250 chips, designed to accelerate AI workloads while reducing costs. Qualcomm's chips are optimized for inference, which is when a trained AI model is used to make predictions from new data. In turn, they are geared toward making generative AI applications more affordable for more businesses. "These innovative new AI infrastructure solutions empower customers to deploy generative AI at unprecedented TCO [total cost of ownership], while maintaining the flexibility and security modern data centers demand," Qualcomm executive Durga Malladi said in a press release. Qualcomm's chips use its Hexagon neural processing unit (NPU) technology, which improves the efficiency of computing tasks and reduces power consumption. The AI200 is due out in 2026. The AI250, which promises improvements in memory bandwidth of up to 10 times current-generation AI accelerators, is expected to launch in 2027. Cloud platform providers are planning to spend over $500 billion to upgrade their AI infrastructure in 2026 alone, according to TrendForce. Much of these investments will be allocated to AI chips, such as those designed by Nvidia and Advanced Micro Devices. Qualcomm wants in on this AI action. The chipmaker is thus making moves to expand beyond its core mobile device market and penetrate the rapidly expanding data center industry. These moves will put Qualcomm in more direct competition with Nvidia and AMD -- two fierce rivals that are not to be taken lightly.

Qualcomm Unveils AI200, AI250 Chips to Challenge NVIDIA in AI Race

Qualcomm Launches AI200 and AI250 Chips, Expands Beyond Smartphones to Target AI Data Centers Qualcomm has announced two new artificial intelligence chips tailored for data centers, marking its expansion beyond the smartphone market. The news pushed Qualcomm shares up by 20%, showing investor optimism about the company's growing AI ambitions. The new chips, dubbed AI200 and AI250, are designed to boost memory capacity and performance and are ideally suited to support large language models, chatbots, and generative AI workloads. The chipmaker also announced that the shipping of AI200 will begin in 2026, and AI250 in 2027.

Qualcomm shares jump 11% as chip giant unveils new AI chips to rival...

Chip giant Qualcomm announced Monday that it will release two new artificial intelligence chips - setting the stage for a head-to-head race with billionaire Jensen Huang's Nvidia. It's a major shift for Qualcomm, which has long focused on manufacturing semiconductors for wireless activity and mobile phones - not for the power-hungry data centers that are powering the AI boom. Shares in Qualcomm jumped 11% Monday to close at $187.68. Qualcomm -- headed by Brazilian-born CEO Cristiano Amon -- said its new, super-powered chips will be available in liquid-cooled server racks, which help customers like cloud service providers save on operational costs. Nvidia and AMD also sell their graphics processing units, or GPUs, in full-rack systems. Large language models like OpenAI's ChatGPT rely on these powerful systems. Still, Qualcomm is facing a steep uphill battle. Nvidia already dominates about 90% of the data center market and boasts a massive $4.6 trillion market cap. Nvidia's chips were used to train OpenAI's ChatGPT, still widely viewed as the most popular AI chatbot. But earlier this month, OpenAI broadened its supply chain. It announced plans to start buying chips from rival AMD, and even potentially take a stake in the company. The AI race has shown no signs of slowing down. Nearly $6.7 trillion in capital expenditures will be spent on AI infrastructure through 2030, according to a McKinsey estimate. Qualcomm's new AI200 and AI250 chips will go on sale in 2026 and 2027, respectively, the company said in a press release. A rack will require 160 kilowatts of power, similar to rack solutions from Nvidia. Qualcomm's chip-based accelerator cards will support 768 gigabytes of memory - more than comparable options from Nvidia and AMD. The company declined to comment on the price of its AI chips, cards or racks. "Qualcomm AI200 and AI250 are designed for frictionless adoption and rapid innovation," Durga Malladi, the firm's senior vice president and general manager for technology planning, said in a statement Monday. "Our rich software stack and open ecosystem support make it easier than ever for developers and enterprises to integrate, manage, and scale already trained AI models on our optimized AI inference solutions." Qualcomm's new AI chips will be based on its Hexagon neural processing units, or NPUs, which are used for its smartphone chips. The company aims to sell its new chips and other parts separately, so customers can design their own racks if they'd like. Nvidia or AMD could even become potential clients for some data center parts, Malladi told reporters last week. "What we have tried to do is make sure that our customers are in a position to either take all of it or say, 'I'm going to mix and match,'" he said.

AI Chip War Just Shifted: Why Memory May Matter More Than Compute | Investing.com UK

When Qualcomm unveiled its new AI200 and AI250 data center accelerators this week, the market saw a familiar story: another chipmaker trying to chip away at Nvidia's stranglehold on AI infrastructure. Qualcomm's stock jumped 22%, its biggest single-day gain in nearly seven years, as investors cheered the news. But beneath the headlines, a more important shift is emerging, one that could upend how investors think about the next phase of the AI boom. The AI hardware race, long dominated by compute horsepower, may be pivoting toward something far less glamorous but potentially more decisive: memory. Qualcomm surges 22% and it is betting that the AI chip race is shifting away from raw compute and toward memory capacity, and its massive LPDDR-based design could give it a real edge in the exploding inference market. For the past two years, Nvidia's GPUs have defined the AI gold rush, their raw computational power making them indispensable for training massive models like GPT-4 and Gemini. But as AI systems move from training to deployment (a phase known as inference), the physics of performance change. Modern AI inference workloads are increasingly memory-bound rather than compute-bound. As models grow in size and context windows expand, the challenge isn't how fast chips can compute. It's how quickly they can feed data to those processors. "It's like having a supercar stuck in traffic," says one industry analyst. "The compute cores are ready to roar, but the data highway is too narrow." This is where Qualcomm's bet looks different. Each of its new AI accelerator cards includes up to 768 gigabytes of LPDDR memory, about 10 times more than Nvidia's H100 GPU configuration. Instead of using expensive High Bandwidth Memory (HBM), Qualcomm is using the low-power LPDDR technology it perfected in smartphones. That shift is more than a cost optimization. It's a rethinking of what matters most in AI hardware. Industry insiders have started calling AI's current bottleneck the "martini straw problem": the compute engine is the glass, but the data flows through a straw. No matter how powerful the chip is, it's limited by how quickly data can move in and out. Qualcomm's approach widens that straw. LPDDR memory (cheaper, denser, and more scalable) offers up to 13 times more capacity per dollar compared to HBM. That makes it possible to run large language models (LLMs) or multimodal AI inference workloads directly in memory, without constant data shuffling. In practice, that means faster responses, lower latency, and much lower energy draw. All major advantages for data centers where power and efficiency are now critical constraints. And the timing is right. According to research from Cambrian AI and Grand View, inference workloads will outnumber AI training by a factor of 100 to 1 by 2030. As AI applications proliferate across devices and industries, memory-rich inference could become the defining performance metric of the decade. For investors, the implications are significant. AI's first boom phase rewarded companies that could deliver maximum compute: Nvidia, AMD, and to a lesser extent, Intel. The next wave will likely reward those who can deliver more efficient, lower-cost inference at scale. Qualcomm's edge lies in combining its mobile DNA (decades of building chips that run complex workloads within strict power and thermal limits) with data center-grade scalability. Independent studies cited by the company show its Cloud AI 100 Ultra architecture consuming 20 to 35 times less power than comparable Nvidia configurations for certain inference workloads. In a world where data center energy consumption is doubling every three years, and utilities are warning of grid constraints, those savings aren't just nice-to-have. They're critical. "If inference becomes the dominant workload, and if power costs continue to climb, efficiency could matter more than peak performance," says one semiconductor analyst. "That's the space Qualcomm knows best." Memory capacity per AI accelerator card for Nvidia H100, Nvidia H200, Qualcomm AI200, and Qualcomm AI250, showing Qualcomm's significant advantage, with AI200 and AI250 having an order of magnitude more memory than Nvidia's top AI GPUs. This lends weight to the argument that memory capacity is becoming a critical bottleneck in AI inference computing. This shift from compute-centric to memory-centric AI could also reshape the competitive landscape. Nvidia's ecosystem remains unmatched for AI training, thanks to its proprietary CUDA software stack and developer lock-in. But inference (where models are deployed and scaled) is far less sticky. It's also far more price-sensitive. That opens the door for companies like Qualcomm to compete on new terms. Rather than building bigger GPUs, Qualcomm is optimizing for the reality that most AI models won't be retrained every day. They'll just need to run efficiently, everywhere. It's a bet that echoes a pattern from the early 2000s, when Intel's dominance in high-performance desktops was gradually eroded by ARM-based chips optimized for mobile computing. Today, that same philosophy (smarter, smaller, cheaper, more efficient) is migrating into the data center. Qualcomm's path has challenges. The company lacks deep relationships in the enterprise data center space, where Nvidia's brand and ecosystem are entrenched. Winning design wins from hyperscalers won't be easy. But Qualcomm isn't going it alone. It's partnering with Saudi-backed Humain, a sovereign AI initiative investing over $40 billion in infrastructure, providing both capital and early deployment scale. If successful, that could offer Qualcomm a way to leapfrog traditional cloud channels and tap into a wave of sovereign AI spending outside U.S. hyperscalers' reach. At a forward P/E ratio of roughly 15-20x (depending on estimates), Qualcomm trades at a significant discount to both Nvidia's 52-57x and AMD's 26-27x. This valuation gap reflects market skepticism about Qualcomm's ability to compete in data centers, but also creates asymmetric upside if the company captures even modest share of the $254 billion inference market projected by 2030. The company's fundamentals remain solid: 2.1% dividend yield backed by a conservative 33% payout ratio, 20 consecutive years of dividend increases, and strong free cash flow supporting both dividends and buybacks (6.7% total shareholder yield). Analysts project 11-15% earnings growth through 2026, with the AI data center entry potentially accelerating that trajectory. The risk, however, exists. Qualcomm faces execution challenges in enterprise sales where it lacks Nvidia's relationships, and the AI200 won't ship until 2026. But for patient investors, the Saudi Humain anchor customer ($40+ billion commitment) and partnerships with emerging sovereign AI initiatives provide revenue visibility that de-risks the data center entry. For small and mid-size investors, the signal is clear: the semiconductor story is evolving. AI's next trillion-dollar opportunity may not come from the next great GPU, but from solving the data bottleneck that makes existing GPUs inefficient. Qualcomm's memory-first design could prove to be the right architecture for that future, a classic case of a challenger seeing the inflection point before the incumbents do. With the AI inference market projected to reach $254 billion by 2030, and the edge AI market another $66 billion, Qualcomm's positioning looks less like a side bet and more like a strategic pivot into the heart of the next computing era. Nvidia still dominates the training race. But the inference marathon is just beginning, and Qualcomm may have chosen the smarter race to run.

Qualcomm stock jumps after unveiling new AI chips to challenge Nvidia By Investing.com

Investing.com -- Qualcomm (NASDAQ:QCOM) stock rose 12% Monday after the company announced its entry into the data center AI market with two new inference-optimized chips designed to compete with Nvidia's (NASDAQ:NVDA) offerings. The semiconductor giant unveiled the Qualcomm AI200 and AI250 chip-based accelerator cards and racks, which the company claims will deliver rack-scale performance and superior memory capacity for generative AI inference workloads at industry-leading total cost of ownership. The AI200 solution features 768 GB of LPDDR memory per card, while the AI250 introduces an innovative near-memory computing architecture that provides over 10x higher effective memory bandwidth with lower power consumption. Both solutions include direct liquid cooling and support for PCIe and Ethernet connectivity. Qualcomm's new offerings come with a comprehensive software stack supporting popular AI frameworks and include features like one-click deployment of Hugging Face models. The company is positioning these products as cost-effective alternatives for data centers running AI inference workloads. The move represents Qualcomm's strategic expansion beyond its traditional mobile chip business into the rapidly growing data center AI market currently dominated by Nvidia. However, commercial availability is still some time away, with the AI200 expected in 2026 and the AI250 in 2027. Qualcomm also committed to an annual cadence for its data center AI roadmap going forward, focusing on inference performance, energy efficiency, and competitive total cost of ownership.

Qualcomm's new AI chips send stock soaring

Qualcomm rose over 11% in New York yesterday, after the company unveiled two new chips, AI200 and AI250, dedicated to artificial intelligence inference in data centers. These solutions offer high performance, large memory capacity, and low operating costs for generative AI applications. The AI200 supports up to 768 GB of memory per card, while the AI250 introduces a new, faster, and more energy-efficient memory architecture. Both models feature liquid cooling, PCIe and Ethernet connections, and enhanced security features. According to Durga Malladi, senior vice president of Qualcomm Technologies, these products will make it easier for companies to deploy large-scale AI models. Commercial launches are planned for 2026 for the AI200 and 2027 for the AI250.

Qualcomm: positioning itself in the artificial intelligence chip market to compete with Nvidia and AMD

On Monday,Qualcomm unveiled its first chips for artificial intelligence data centers, the AI200 and AI250, which are set to launch in 2026 and 2027, respectively. This strategic shift marks the mobile semiconductor specialist's entry into a segment dominated by Nvidia and AMD. These new accelerators, designed to operate in parallel in complete liquid-cooled systems, are aimed at infrastructures hosting the most powerful AI models. They are based on Hexagon neural processing units (NPUs) already used in Qualcomm smartphones, but adapted here to an industrial scale. Following the announcement, Qualcomm's share price rose by nearly 9% during trading. The group intends to focus on the inference phase, i.e., the execution of AI models, promising better energy cost-efficiency than its competitors. Each rack would consume around 160 kilowatts, a level comparable to Nvidia systems, while offering operating savings. Qualcomm is also adopting a modular approach: its customers will be able to purchase either a complete system or components to integrate into their own architectures. This flexibility could appeal to cloud giants such as Amazon, Alphabet, and Microsoft, which are looking to diversify their suppliers. According to McKinsey, global investment in data centers is expected to reach $6.7 trillion by 2030, with the majority of that dedicated to AI. Qualcomm hopes to capitalize on this momentum to gain ground against Nvidia, which still holds over 90% of the GPU market. In May, the company signed an agreement with Saudi Arabia's Humain to deploy its inference chips in regional infrastructures totaling up to 200 megawatts of power. Finally, Qualcomm points out that its AI cards, capable of handling up to 768 GB of memory, offer substantial gains in energy consumption and memory performance compared to competing solutions.

Qualcomm accelerates data center push with new AI chips launching next year

(Reuters) -Qualcomm on Monday unveiled two artificial intelligence chips for data centers, with commercial availability from next year, as it pushes to diversify beyond smartphones and expand into the fast-growing AI infrastructure market. Shares of Qualcomm surged nearly 15% on the news. The new chips, called AI200 and AI250, are designed for improved memory capacity and running AI applications, or inference, and will be available in 2026 and 2027, respectively. Global investment in AI chips has soared as cloud providers, chipmakers and enterprises rush to build infrastructure capable of supporting complex, large language models, chatbots and other generative AI tools. Nvidia chips, however, underpin much of the current AI boom. Qualcomm, to strengthen its AI portfolio, agreed to buy Alphawave in June, which designs semiconductor tech for data centers, for about $2.4 billion. In May, Qualcomm also said it would make custom data center central processing units that use technology from Nvidia to connect to the firm's artificial intelligence chips. Qualcomm said the new chips support common AI frameworks and tools, with advanced software support, and added they will lower the total cost of ownership for enterprises. The San Diego-based company also unveiled accelerator cards and racks based on the new chips. Earlier this month, peer Intel announced a new artificial intelligence chip called Crescent Island for the data center that it plans to launch next year. (Reporting by Harshita Mary Varghese in Bengaluru; Editing by Vijay Kishore)