Community

Featured post
Sparsity-based model compression from the Neuronix AI Labs acquisition promises up to 2x faster CNN inference on low-power, SEU-immune hardware.
Featured post
The USC-run service that pioneered multi-project wafers in 1981 now offers a path from shared shuttle runs to full production on GlobalFoundries' FinFET and FD-SOI processes.
Featured post
IBM leveraged its nanostack architecture, a technique that "builds a chip like a city," to pack nearly 100 billion transistors on a fingernail-sized chip.
Featured post
Called Gleanmer, the 16-nm system-on-chip swaps memory-hungry voxels for compact Gaussian “blobs” to map and navigate at roughly 6 mW.
Featured post
The Korean unicorn packages four NPU dies with 144 GB of HBM3E and bets on an open software stack to serve large language models at lower cost per token.
Featured post
Lotus Microsystems, Oriole Networks, and Atomera have each rolled out technologies tackling the physical limits slowing AI infrastructure.
Featured post
The new XpressConnect PCIe 6.0 and CXL 3.1 Retimers reign in signal integrity challenges in AI data centers and other high-bandwidth installations.
Featured post
Two new CoolGaN BDS additions replace back-to-back silicon MOSFET pairs with chip-scale packages.
Featured post
Broadcom’s high-integration Wi-Fi 8 & NPU-accelerated 50G PON gateway SoCs aim to build a cohesive, 50 Gbps broadband access ecosystem for the AI-infused home.
Featured post
The 64-MHz Cortex-M0+ part includes single-shunt FOC support, dual rail-to-rail op amps, and four programmable comparators in packages as small as QFN20.
Top