NVIDIA’s Rubin AI Platform Alone Will Devour More LPDDR Memory in 2027 Than Apple and Samsung Combined, Starving Smartphone Supply

May 17, 2026 at 10:15am EDT
LPDDR Memory Demand For NVIDIA Rubin In AI To Exceed The Combined Demand of Samsung & Apple In 2027

LPDDR memory demand in the AI & server segment is going to outstrip smartphone demand massively in 2027, with next-gen platforms such as NVIDIA Rubin and AMD MI400 requiring way more DRAM to meet the needs of AI firms.

AI LPDDR Demand Is Going To Push Back The Smartphone Market Massively, NVIDIA Rubin's Demand Alone To Exceed Both Samsung & Apple

Incoming memory demand from Agentic AI use cases is going to be insane. We know that DRAM is going to retain its crucial status in powering upcoming AI servers, and future platforms are going to house more capacity as compute requirements grow.

Related Story Apple’s AR Glasses To Replace The Vision Pro Lineup For Its Mass Market Appeal, But Display-Equipped Spectacles Still Several Years Away

As per Citrini Research, it is estimated that NVIDIA's Rubin AI platform alone is going to outstrip both Samsung's and Apple's DRAM demand in 2027. Rubin is just one platform, and the same can be seen across upcoming hyperscalers, where LPDDR demand has grown exponentially as model sizes continue to grow.

In 2027, NVIDIA's Rubin AI platforms are estimated to gobble up over 6000 Million GBs worth of LPDDR memory. LPDDR memory is the go-to standard for AI servers as its high capacity, compact layout, & low-power characteristics make it ideal for such applications. For 2027, Apple is estimated to be the leading LPDDR DRAM buyer for its iPhones with a demand of 2966 Million GBs, followed by Samsung, which has an estimated demand of 2724 Million GBs in LPDDR DRAM.

Both Apple and Samsung will have a combined (Estimated) demand of 5720 Million GBs worth of DRAM. This means that NVIDIA's Vera Rubin AI racks will have a 6% higher demand.

Meanwhile, AMD's MI400 & other Agentic AI-aimed platforms will further raise the demand for LPDDR memory. Currently, LPDDR5 & LPDDR5X solutions are being optimized for upcoming servers, with Micron already introducing its 256 GB LPDDR5X SOCAMM2 memory. SK Hynix is also mass producing 192 GB LPDDR5X SOCAMM2 memory for NVIDIA's Vera Rubin platform.

On the AMD front, the AI-Specific Verano CPUs and MI455X chips for Helios racks will also feature support for LPDDR5X memory, further straining the DRAM supply chain. Samsung, SK Hynix, & Micron have gone all-in on the latest LPDDR standards, boosting their production to meet the demand of the next-generation AI ecosystem. At the same time, this massive demand will likely not be fulfilled with the current production capacities; as such, newer facilities are being constructed to narrow down the supply-demand gap.

For smartphone makers, this huge LPDDR demand from the AI segment will cause many obstacles, such as product shortages and price increases, which have already affected various segments of the tech world.

News Source: @jukan05

About the author: A Software Engineer by training and a PC enthusiast by passion, Hassan Mujtaba serves as Wccftech's Senior Editor for hardware section. With years of experience in the industry, he specializes in deep-dive technical analysis of next-generation CPU and GPU architectures, motherboards, and cooling solutions. His work involves not only breaking news on upcoming technologies but also extensive hands-on reviews and benchmarking.

Follow Wccftech on Google to get more of our news coverage in your feeds.