Newsroom | 43145 results

WASHINGTON & WAKEFIELD, Mass.--(BUSINESS WIRE)--Wireless Power Consortium and NFC Forum Issue Joint Statement on Industry Standard Development...

VALHALLA, N.Y.--(BUSINESS WIRE)--FUJIFILM North America Corporation, Electronic Imaging Division, announces the launch of its FUJINON XF23mmF2.8 R WR (XF23mmF2.8 R WR) lens. Designed with portability in mind, XF23mmF2.8 R WR was developed to be thin and unobtrusive while delivering powerful resolution and autofocus (AF) performance. “The first word that comes to mind when thinking of XF23mmF2.8 R WR is ‘versatility’,” said Victor Ha, vice president, Electronic Imaging and Optical Devices Divisi...

ORLANDO, Fla.--(BUSINESS WIRE)--Scalable Omni-Channel software platform with a single app leverages over 7 billion existing smartphones to reach a global audience and drive revenue....

VALHALLA, N.Y.--(BUSINESS WIRE)--FUJIFILM North America Corporation, Electronic Imaging Division, announces the launch of FUJIFILM X-E5 (X-E5) rangefinder-style mirrorless digital camera. X-E5 combines the classic look and feel that has made Fujifilm’s X-E series a sought-after tool among photographers, with an updated design and the latest features X Series has to offer including a customizable Film Simulation (FS) dial, in-body image stabilization, enhanced viewfinder capabilities, and more....

KARLSRUHE, Germany--(BUSINESS WIRE)--Cinemo, a global leader in high-performance and automotive-grade multimedia playback, streaming, media management, connectivity, and cloud middleware, is pleased to announce that India’s leading SUV manufacturer Mahindra has chosen Cinemo as its infotainment solutions technology provider. Cinemo is set to power bring-your-own-device (BYOD) infotainment capabilities for Mahindra’s range of electric origin SUVs (eSUVs), delivering exceptional infotainment expe...

中國上海--(BUSINESS WIRE)--(美國商業資訊)-- 芯原股份（芯原，股票代號：688521.SH）今日宣布其超低能耗且高效能的神經網路處理器（NPU）IP現已支援在行動端進行大語言模型（LLM）推理，AI算力可擴充至40 TOPS以上。該高能效NPU架構專為滿足行動平台與日俱增的生成式AI需求而設計，不僅能夠為AI PC等終端設備提供強勁算力支援，而且能夠因應智慧型手機等行動終端對低能耗更為嚴苛的挑戰。 芯原的超低能耗NPU IP具備高度可設定、可擴充的架構，支援混合精確度運算、稀疏化最佳化和並行處理。其設計融合了高效的記憶體管理與稀疏感知加速技術，顯著降低運算負載與延遲，確保AI處理流暢、回應迅速。該NPU支援數百種AI演算法，如AI雜訊抑制（AI-NR）和AI超分（AI-SR）等，並與Stable Diffusion和LLaMA-7B等主流AI模型相容。同時，該NPU IP還可與芯原其他處理器IP順暢整合，實現異構運算，協助SoC設計者打造滿足多元化應用需求的AI解決方案。 此外，芯原的超低能耗NPU IP還支援TensorFlow Lite、ONNX和PyTorc...

XANGAI--(BUSINESS WIRE)--A VeriSilicon (688521.SH) anunciou hoje que sua Unidade de Processamento de Rede Neural (NPU) IP com consumo extremamente baixo de energia e alto desempenho agora suporta inferência de grandes modelos de linguagem (LLMs) em dispositivos, com escalabilidade de desempenho de computação de IA além de 40 TOPS. Esta arquitetura de NPU com eficiência energética foi criada especificamente para satisfazer a crescente demanda por recursos de IA generativa em plataformas móveis....

SHANGHÁI--(BUSINESS WIRE)--VeriSilicon (688521.SH) anunció hoy que su IP de Unidad de Procesamiento de Redes Neuronales (Neural Network Processing Unit, NPU) de consumo ultrabajo y alto rendimiento admite ahora la inferencia en el dispositivo de modelos de lenguaje de gran tamaño (LLM) con un rendimiento de cálculo de IA que supera los 40 TOPS. Esta arquitectura NPU de bajo consumo está diseñada específicamente para satisfacer la creciente demanda de capacidades de IA generativa en plataformas...

SHANGHAI--(BUSINESS WIRE)--VeriSilicon (688521.SH) a annoncé aujourd’hui que son unité de traitement de réseau neuronal (NPU) ultra-basse consommation et hautes performances prend désormais en charge l’inférence intégrée de grands modèles de langage (LLM), avec des performances de calcul d’IA dépassant 40 TOPS. Cette architecture NPU écoénergétique est spécialement conçue pour répondre à la demande croissante de capacités d’IA générative sur les plateformes mobiles. Elle offre non seulement des...

SHANGHAI--(BUSINESS WIRE)--VeriSilicon (688521.SH) ha annunciato oggi che il suo IP NPU (Neural Network Processing Unit) a bassissima energia e ad alte prestazioni supporta ora l'inferenza on-device di grandi modelli linguistici (LLM) con prestazioni di calcolo IA che vanno oltre i 40 TOPS. Questa architettura NPU a efficienza energetica è stata appositamente sviluppata per soddisfare la domanda in crescita di capacità di IA generativa sulle piattaforme mobili. Non solo offre potenti prestazion...