Sources: TSMC is considering expanding into manufacturing 3D DRAM at the request of Nvidia, its biggest customer, amid memory chip shortages
TSMC considers 3D DRAM manufacturing expansion at Nvidia's request to address memory chip shortages.
Search the full wire by company, model, lab, or keyword. Every story we have ever aggregated.
TSMC considers 3D DRAM manufacturing expansion at Nvidia's request to address memory chip shortages.
Nvidia requires Southeast Asian Cloud Partner applicants to enroll in chip location tracking to combat semiconductor smuggling.
Chinese AI chipmakers Moore Threads, MetaX, Iluvatar CoreX report Q3 profitability amid US export controls on Nvidia, but face supply chain and memory cost challenges.
US lawmakers request briefing with Nvidia CEO Jensen Huang over alleged ties to executive accused of passing intelligence to China.
US gubernatorial races in Texas, Ohio, Georgia, Pennsylvania may slow AI data-center expansion; Taiwan suppliers and Nvidia monitor regulatory shifts.
South Korea's JNTC joins Nvidia supplier list, providing through-glass via substrate samples for advanced chip packaging.
OpenAI and Nvidia adjust US data center expansion strategies following election-driven community backlash.
Nvidia launches Kubernetes Engine (NVKE), a managed orchestration platform for GPU workloads and AI inference clusters.
Naver launches Nvidia-backed AI computing services business in South Korea after securing US$9B Brookfield infrastructure deal.
Nvidia reportedly approached Apollo about converting Brightspeed's 130+ buildings into small data centers for chip deployment.
~600k Nvidia H200 chips stuck in Hong Kong as Beijing delays mainland import approval.
Also covered by DIGITIMES
XStack is a benchmark evaluating agents on migrating ML repositories between PyTorch/NVIDIA and JAX/TPU ecosystems across hardware and frameworks.
Foxconn integrates Nvidia DSX POD containerized data center systems using modular manufacturing for rapid deployment.
One underperforming GPU can slow an entire distributed training job. The node may pass basic health checks and show normal utilization, leaving operators to... One underperforming GPU can slow an entire distributed training job. The node may pass basic health checks and show…
In an AI factory, high bandwidth memory (HBM)is the fastest, most expensive, and most contested memory tier. Every byte spent on a redundant weight copy, an... In an AI factory, high bandwidth memory (HBM)is the fastest, most expensive, and most contested memory tier. Every…
Box integrates Nvidia Nemotron into content retrieval agent, claiming 4x faster queries, 20% fewer errors, 90% token reduction.
Moonshot AI circumvents U.S. export controls by leasing Nvidia Blackwell chips through European cloud providers.
Also covered by DIGITIMES
Moreh claims AMD MI355X achieves 2x+ inference throughput versus Nvidia B200 in independent benchmarking.
Despite AI skepticism, server supply chain reports strong orders; Nvidia transitions platforms while ASICs fill gap, Vera Rubin ramping Q1 2027.
oToBrite and Yuan integrate 3D stereo vision camera module with Nvidia Jetson Thor for robot and autonomous vehicle perception.
Nvidia DSX POD reference design enables prefabricated, modular AI data center delivery to accelerate deployment and reduce build times.
As GPU clusters scale, operations span compute, network, hardware, and scheduling. Site reliability engineers (SREs) often have to examine multiple dashboards... As GPU clusters scale, operations span compute, network, hardware, and scheduling. Site reliability engineers (SREs)…
As artificial intelligence and accelerated computing continue to push the boundaries of what's possible, the latest NVIDIA CUDA release arrives to unlock... As artificial intelligence and accelerated computing continue to push the boundaries of what’s possible, the latest…
Turn production evidence into prioritized issues and custom evaluations that help you improve prompts, tools, models, memory, skills, and harness code. When you... Turn production evidence into prioritized issues and custom evaluations that help you improve prompts, tools…
Thompson analyzes how US midterm election results affect AI infrastructure policy, data center permitting, and competitive dynamics vs. China.
Nvidia RTX Spark encrypted firmware reduces Linux inference speed to 35% of baseline, impacting ML workload performance.
Samsung to produce HBM base dies for AMD and Nvidia at Texas foundry, with AMD as primary customer.
Chinese AI firms (Alibaba, Tencent, ByteDance) ordered 6M Nvidia RTX Pro 5500 chips worth $70B; expect ~1% delivery within 4 months.
DeepSeek plans to train V5 model on Huawei Ascend 950 chips in 2027, extending DeepSeek-Huawei partnership from inference to full training.