LPDDR5x
PulseAugur coverage of LPDDR5x — every cluster mentioning LPDDR5x across labs, papers, and developer communities, ranked by signal.
3 day(s) with sentiment data
-
Qualcomm validates stacked AI memory architecture for 2027 accelerators
Qualcomm has successfully validated its High Bandwidth Compute (HBC) memory architecture, a key component for its upcoming Dragonfly AI250 data center accelerators. This stacked memory design places DRAM directly on top…
-
Sipeed launches $3,999 AI rival to Nvidia DGX Spark and AMD Ryzen AI Halo
Sipeed has unveiled the MaixStation, a new AI workstation designed to compete with Nvidia's DGX Spark and AMD's Ryzen AI Halo. This Chinese-made system features Axera's AX8860P processor, paired with 128GB of LPDDR5X me…
-
NVIDIA launches RTX Spark chip to power AI-focused Windows laptops
NVIDIA has launched its new RTX Spark system-on-a-chip (SoC) designed for Windows PCs, aiming to compete with Apple Silicon and other high-end processors. The RTX Spark integrates NVIDIA's Grace Arm CPU cores and Blackw…
-
AI as a Skillful Tool and Intel's New GPU Accelerator
The first item discusses how Artificial Intelligence is a tool that requires skill to use effectively, emphasizing the human element in its application. The second item details Intel's new D270 GPU AI Accelerator, highl…
-
Intel launches D270 GPU AI Accelerator with 480GB memory · 2 sources tracked
Intel has launched its D270 GPU AI Accelerator, designed for data center applications. This new accelerator boasts up to 480 GB of LPDDR5X memory, aiming to enhance AI processing capabilities. The product is part of Int…
-
Intel GPU D-Series: Crescent Island accelerator specs revealed
Intel has officially named its Crescent Island AI accelerator the GPU D-Series and revealed key performance specifications. The chip will deliver 1.3 PFLOPS of FP8 compute and 1.5 TB/s of memory bandwidth, supporting up…
-
GMKtec launches EVO-X3 mini-PC with AMD Ryzen AI Max+ for local AI tasks
GMKtec has released the EVO-X3, a new mini-PC featuring AMD's Ryzen AI Max+ 395 processor. This compact desktop is designed for AI development and local inference tasks, offering up to 128GB of LPDDR5X memory. The EVO-X…
-
AMD readies new EPYC Verano AI host CPU with dedicated SB1 socket
AMD is reportedly developing a new EPYC Verano CPU specifically for AI servers, slated for release in the second half of 2027. This processor will utilize a new, smaller SB1 socket, distinguishing it from other Zen 6 EP…
-
Lenovo Googlebook 15 leak reveals premium specs and AI features
Lenovo is reportedly developing a premium laptop, the Googlebook 15, which is expected to feature high-end specifications. The device is rumored to include a 15.3-inch 120Hz OLED display with a 2,880 x 1,800 resolution …
-
Anthropic blocks surveillance AI accounts; YC avoids distillation crackdown; Positron AI raises $875M for chip
Anthropic has blocked AI accounts linked to surveillance operations in Mali, China, and Iran, though some systems remain active. Meanwhile, Y Combinator's Garry Tan has decided against taking action on AI model distilla…
-
CXMT begins mass production of LPDDR6 mobile DRAM
ChangXin Memory Technologies (CXMT) has commenced mass production of LPDDR6 mobile DRAM. The new memory features 16Gb dies, offers packages up to 16GB, and achieves a peak data rate of 12,800Mbps. This advancement provi…
-
Volatile vs. Non-Volatile Memory: Understanding the Core Differences
Volatile memory, such as DRAM and SRAM, requires a constant power supply to retain data, making it suitable for active processing and CPU caches. In contrast, non-volatile memory, like NAND flash used in SSDs and NOR fl…
-
China's CXMT announces LPDDR6 mass production, challenging Western chipmakers
China's ChangXin Memory Technologies (CXMT) has announced the mass production of LPDDR6 memory, a significant advancement that places them on par with Western chipmakers like Samsung, SK Hynix, and Micron. This new memo…
-
Intel's Crescent Island GPU prioritizes memory capacity for AI inference
Intel has introduced its new Crescent Island GPU, which prioritizes memory capacity over raw speed. This design choice utilizes LPDDR5X memory, allowing for up to 480GB of capacity on a single 350W PCIe card. This appro…
-
Samsung integrates logic into LPDDR5X memory for faster AI inference
Samsung has developed LPDDR5X-PIM, a new memory technology that integrates logic units directly into the memory chips. This Processing-in-Memory (PIM) approach aims to reduce the cost and power consumption associated wi…
-
Intel unveils practical AI accelerator Crescent Island for inference
Intel has revealed new details about its Crescent Island AI accelerator, built on the Xe3P architecture, at the Hot Chips symposium. This accelerator is designed for inference tasks, featuring a 350W air-cooled PCIe car…
-
AI servers adopt mobile memory, driving up costs and straining supply
AI servers are increasingly incorporating LPDDR5X memory, traditionally used in mobile devices, due to rising CPU core counts and escalating data demands in AI and supercomputing workloads. This shift is driven by LPDDR…
-
Nvidia RTX Spark N1X prototype Surface Laptop Ultra shows early promise, faces driver issues
A prototype Microsoft Surface Laptop Ultra, reportedly featuring an unreleased Nvidia RTX Spark N1X System on Chip (SoC), has been put through preliminary testing by a tech enthusiast. The N1X SoC is designed for AI tas…
-
Micron unveils low-power DRAM for AI data centers
Micron has developed a new low-power DRAM, LPDDR5X, which consumes approximately one-third the power of DDR5 for AI data center workloads. This new memory technology accounts for less than 7% of total system power, pote…
-
Nvidia Vera CPU targets agentic AI with custom Olympus cores · 4 sources tracked
Nvidia has revealed new details about its Vera CPU, designed to enhance agentic AI workloads. The company claims Vera offers significant improvements in single-threaded performance and memory bandwidth compared to tradi…