Why Does Phone Memory Cost More Than HBM After Entering NVIDIA Cabinets?
NVIDIA Vera Rubin Ultra cabinets will be configured with 216TB of LPDDR5X memory, with a total cost of approximately $2.8 million, exceeding the $2.5 million cost of 124.4TB of HBM4E in the same cabinet.
Key Takeaways
- Key Highlight：NVIDIA Vera Rubin Ultra cabinets will be configured with 216TB of LPDDR5X memory, with a total cost of approximately $2.8 million, exceeding the $2.5 million cost of 124.4TB of HBM4E in the same cabinet.
- Innovation & Tech：Highlights advancements in NVIDIA, Why, Does, demonstrating rapid progress in model capabilities.
- Industry Impact：Reported via 雷峰网 (CN), offering actionable signals for developers and technology leaders.
NVIDIA's new generation AI server architecture is changing memory configuration strategies. LPDDR5X memory, originally primarily used for mobile devices, is being introduced on a large scale into the Kyber cabinets of the Vera Rubin Ultra NVL144, and its total cost exceeds that of traditional HBM memory.
HBM has always been the core memory for AI training and inference, but the introduction of LPDDR5X may indicate new considerations for balancing memory capacity and bandwidth requirements. This architectural adjustment reflects the profound impact of AI compute demand on the hardware supply chain, prompting the migration of traditional consumer electronics components to data centers.
This change will reshape the cost structure of AI servers. Large-scale demand for LPDDR5X may drive up its price and affect the memory supply for consumer electronics such as smartphones. At the same time, this also provides new growth space for memory vendors, accelerating the hardware iteration of AI infrastructure.
The boundaries of AI compute hardware are becoming blurred, and the lines between traditional components and dedicated chips are becoming increasingly clear. Future AI server design will focus more on optimizing overall efficiency and cost, rather than relying solely on single high-performance components.
This page provides an editorial summary based on publicly available information. It is not a republished article. Use the source link below for the original report.
Industry Insights & Analysis
As artificial intelligence rapidly evolves, breakthroughs surrounding NVIDIA, Why, Does, Phone are shifting toward scalable, robust real-world implementations.
Driven by both open-source ecosystems and proprietary model architectures, the integration between compute optimization, data engineering, and agentic workflows is accelerating. This development provides a strategic benchmark for upcoming AI tooling and developer workflows.