[CLSTR]
< Back to situations

Monitor this situation.

[SITUATION] · [QUIET] · [TECHNOLOGY]

2 clusters · 7 sources · 3 days · First seen · Last updated

AI hardware and semiconductor development

Overview

The semiconductor industry is advancing specialized hardware to support artificial intelligence and large-scale model inference.

Micron is developing Edge Inference Memory (EIM), a high-performance DRAM intended for physical AI applications including autonomous driving and humanoid robotics. Concurrently, NVIDIA has launched RTX Spark, an Arm-based platform for Windows PCs using the N1X System-on-Chip. This architecture combines a Grace CPU with a Blackwell GPU to support AI models exceeding 32GB on local devices.

Intel has also introduced its GPU D-series, internally named ‘Crescent Island’. Utilizing the Xe3P architecture, these data center GPUs are designed for AI inference tasks, capable of handling models with up to 300 billion parameters on a single card. Intel is also showcasing new rack configurations powered by Xeon processors that follow the Open Rack v3 standard.

Entities

Open Compute Project · Micron · GRACE · Arm · Nvidia

Timeline

  1. [TECHNOLOGY] 3 sources
    Intel unveils GPU D-series for AI inference

    Intel announced its GPU D-series for AI inference, featuring up to 480GB of memory and support for models with over 1 trillion parameters using a standard PCIe form factor.

  2. [TECHNOLOGY] 4 sources
    Micron and NVIDIA announce new hardware for AI and humanoid robotics

    Micron is developing Edge Inference Memory for humanoid and physical AI, while NVIDIA has launched its Arm-based RTX Spark platform for Windows PCs to support large-scale local AI processing.

Sources

finance.technews.tw · global-net.co.jp · ithome.com · meeco.kr · thebridge.jp · thelec.kr · zdnet.co.kr