Memory– tag –
-
Qualcomm Unveils Next-Gen Hexagon NPU with 50% Memory Increase for Accelerated Agentic AI, Potentially in Galaxy S27 Ultra
SammyGuru USA Overview Qualcomm announced its next-generation Hexagon NPU for upcoming Snapdragon platforms, significantly enhancing the speed and capability of agentic AI. This new AI hardware, expected to be featured in devices like th... -
New LLM Inference Algorithm for Compute-in-Flash Systems Achieves 15x KV Cache Traffic Reduction, Delivering Energy and Latency Savings for Llama-3.1-8B and Qwen-2.5-7B
arXiv Unknown Overview Research published on arXiv introduces a novel algorithm enabling LLM inference on Compute-in-Flash systems, mitigating memory bandwidth limitations. This method employs end-to-end integer-only quantization and a d... -
Samsung Research Unveils AnySimLite: Sub-700KB, Sub-30ms On-Device AI Matching 7B-Parameter LLMs for Speech Classification
Samsung Research South Korea Overview Samsung Research has introduced AnySimLite, a lightweight few-shot similarity encoder capable of delivering state-of-the-art performance for multiple on-device speech-adjacent classification tasks. O... -
Nanoimprint Lithography Poised to Break Manufacturing Barriers in Nano-Optics, Highlighted at CIOE 2026
CIOE (China International Optoelectronic Expo) China Overview The 2026 "Nanoimprint Manufacturing Technology Forum" at CIOE will spotlight Nanoimprint Lithography (NIL) as a groundbreaking technique for nano-optics, offering high resolut... -
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 ... -
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... -
Optalysys Demonstrates Photonic Hardware for Compute-in-Transit, Revolutionizing AI and Cryptographic Workloads by Eliminating Data Movement Bottlenecks
Quantum Zeitgeist UK Overview Optalysys has demonstrated innovative photonic hardware capable of performing computations directly as data travels from storage. This 'in-transit computing' model addresses the critical data movement bottle... -
Arteris Reports High Premium for AI Chip IP, Reflecting Soaring Demand for Specialized Silicon
Design And Reuse USA Overview Arteris has observed a significant premium for AI chip intellectual property (IP) in the market, underscoring the escalating value of specialized IP in AI hardware development. Semiconductor manufacturers ar...
