Install
AI/HPC Accelerators
GPU pods, tensor ASICs, high‑bandwidth memory, and interconnects.
- 4 Tracked terms
- Last 30 days Feed window
What this topic collects on
An article joins this feed when it matches these terms. Each one is also a search of its own.
Related topics
Latest in AI/HPC Accelerators
NHN Cloud Unveils GPU Infrastructure at AI FESTA
2+ hour, 58+ min ago (405+ words) NHN Cloud (hereinafter NHN Cloud) is participating in AI FESTA 2026 (hereinafter AI FESTA) to showcase a GPU cluster featuring 4,080 B200 GPUs alongside its AI full-stack brand, NHN FactoryX. The cluster ranked 1st in South Korea and 20th globally in the TOP500 supercomputer performance rankings....
From $1.7 Trillion to $3.2 Trillion: Semiconductors’ Next Four Years
10+ hour, 20+ min ago (263+ words) Prices also remain high overall, though not uniformly so. DRAM and NAND prices have barely declined over the past week, and renting an Nvidia B200 accelerator costs approximately $5.72 per hour, down slightly from a March peak of $6.10. Computing power remaining this…...
GIGABYTE AI TOP ATOM 64GB Unified Memory Version Expands Possibilities for Desktop AI Development
2+ hour, 41+ min ago (410+ words) Thailand Business News GIGABYTE AI TOP ATOM 64GB Unified Memory Version Expands Possibilities for Desktop AI Development TAIPEI, Oct. 5, 2026 /PRNewswire/ — GIGABYTE today announced the launch of a new 64GB unified memory version of GIGABYTE AI TOP ATOM, expanding its existing 128GB offering. Available…...
Global Semiconductor Sales Top $1 Trillion Through August
2+ hour, 49+ min ago (174+ words) Circuits Assembly - Global Semiconductor Sales Top $1 Trillion Through August WASHINGTON -- Global semiconductor sales surpassed $1 trillion through August for the first time, according to the Semiconductor Industry Association (SIA). Worldwide chip sales totaled $159.7 billion in August, up 8% from July and 144.3% from…...
Kepler Computing to Begin Memory Production Next Year to Ease AI Chip Bottlenecks
6+ hour, 3+ min ago (203+ words) Kepler expects to produce AI chips with partners including GlobalFoundries without relying on extreme ultraviolet lithography to reach advanced nodes. Chief executive Debo Olaosebikan told EE Times that the company resembles established memory makers such as Micron, SK Hynix, or…...
Kioxia CM9-R 15.36TB E3.S NVMe SSD Review
7+ hour, 19+ min ago (225+ words) Today we are looking at the Kioxia CM9-R 15.36TB E3.S PCIe Gen5 SSD. Compared with the 7.68TB CD9P-R we tested earlier we expect similar read speeds, but the CM9-R should offer better write speeds. As a result, it seems like something fun to…...
AI’s massive appetite could make cheap smartphones much harder to find
11+ hour, 11+ min ago (25+ words) The booming demand for AI infrastructure is putting pressure on the same memory supply used by smartphones, and budget devices are feeling the squeeze first....
A message from John Furrier, co-founder of SiliconANGLE:
11+ hour, 4+ min ago (527+ words) UPDATED 12:46 EDT / OCTOBER 05 2026 Agentic AI infrastructure must support the work between a model’s decisions as agents move from answering questions to executing tasks. Central processing units handle much of that execution, while graphics processing units power model reasoning. Those complementary…...
Broadcom-Samsung Chip Pricing Deal Matches Nvidia Terms [2026]
16+ hour, 47+ min ago (477+ words) South Korean outlet Sedaily reported on July 27, 2026, that Broadcom’s own stated rationale was blunt: TSMC capacity was too costly and too slow for the volumes Broadcom needed. That single framing explains why a company that has built its AI business…...
Tencent scores 100,000 offshore AI chip deal with Oracle for $7 billion despite climbing prices — per-hour costs estimated to be 43% under standard H100 rental rates
14+ hour, 30+ min ago (208+ words) Neither company has said which chips Tencent is renting. Without specifics on the hardware being rented, it is not possible to determine whether this is a bulk deal or if the leasing costs are closer to the actual hardware cost....