GPU– tag –
-
Lumentum Secures Multi-Billion Dollar 1.6T Optical Module Deal with NVIDIA, Coherent Achieves Record Revenue; AI Data Center Optics Investment Accelerates, TSMC to Mass Produce 1.6nm Process
Troy Technical (note.com) Global Overview A weekly report to August 28, 2026, reveals accelerating investment in optical communications driven by AI data center demand. Lumentum secured a multi-billion dollar long-term contract with NVID... -
Marvell Expands Photonic Fabric with Google TPU Ecosystem Partnership, 8 Tbps Optical I/O Chiplet Demo, and Celestial AI Acquisition
FUNDA | Substack USA Overview Google and Marvell announced a strategic partnership for custom silicon programs connecting to the TPU ecosystem, leading Marvell to significantly increase its FY27 revenue outlook for scale-up optics. Ayar ... -
TSMC Predicts Silicon Photonics to Dominar Over Half of Optical Transceiver Market by 2027, Driven by Exploding AI Infrastructure Demand
IN Electronics & Design Taiwan Overview TSMC forecasts that silicon photonics will account for over 50% of the optical transceiver market by 2027, propelled by the accelerated adoption of optical links in AI infrastructure. Their CO-UPE ... -
Liquid Cooling Becomes Essential for AI Data Centers Amid Surging Demand, Addressing Over 130kW Power Consumption of NVIDIA GB200 NVL72 Racks
BIS Research International Overview The explosive growth of AI data centers is rapidly accelerating demand for liquid cooling technologies. High-density GPU clusters, such as NVIDIA's GB200 NVL72 rack consuming over 130kW, generate immen... -
Intel Unveils Next-Gen Processors for Agentic AI at Hot Chips 2026: Featuring 256-Core Diamond Rapids CPU and Inference-Optimized Crescent Island GPU
TechRadar USA Overview Intel showcased its next-generation hardware architectures optimized for agentic AI workloads at Hot Chips 2026. The lineup includes the "Wildcat Lake Core Series 3 SoC" for client and edge devices, the "Crescent I... -
CoreWeave Secures $5 Billion Funding to Expand Global GPU-Accelerated Cloud Infrastructure, Addressing Persistent AI Compute Resource Scarcity
TechCrunch USA Overview CoreWeave, an AI cloud infrastructure provider, announced a massive $5 billion funding round led by a consortium of sovereign wealth funds and private equity firms. This capital injection is earmarked for signific... -
Cerebras Systems Achieves AI Inference Breakthrough with Wafer-Scale Engine 3: Delivering Ultra-Low Latency and High Throughput for Generative AI
Cerebras Systems Press Release USA Overview Cerebras Systems has announced a significant leap in AI inference capabilities with its Wafer-Scale Engine 3 (WSE-3), demonstrating record-breaking low latency and high throughput for large-sca... -
NVIDIA Launches Blackwell Ultra GPU: Doubles FP8 Performance, Achieves 16 TB/s HBM4 Bandwidth, Supports 100,000+ GPU Clusters for Trillion-Parameter AI Training
NVIDIA Corporate Newsroom USA Overview NVIDIA has unveiled its Blackwell Ultra GPU, engineered to meet the extreme demands of training trillion-parameter AI models. This new architecture delivers a 2x increase in FP8 performance and inte... -
Alibaba Develops RISC-V Chip Capable of Running 27 Billion Parameter AI Model Without GPU
Design And Reuse China Overview Alibaba has developed a RISC-V-based chip that can execute a 27 billion parameter AI model without relying on a GPU. This achievement signals significant progress in specialized alternative chip architectu... -
Hot Chips 2026: South Korea’s BOS Semiconductors Unveils ‘Eagle-N’ AI Accelerator with up to 2,000 TOPS for Physical & Automotive AI
The Elec Inc. South Korea Overview At Hot Chips 2026, South Korean BOS Semiconductors introduced 'Eagle-N,' an AI accelerator specifically designed for physical and automotive AI applications. This innovative chip boasts a data-efficient...
