In Entwicklung
Business·5 sa önceKI-Zusammenfassung
建通AI高階材料布局有成 輝達GPU液冷散熱銅材獲進展
銅材廠建通積極布局AI高階材料供應鏈，針對輝達下一代GPU液冷散熱需求開發的微通道散熱銅材已取得進展，預計今年第四季至明年初開始出貨，看好AI產品放量帶動下，明年營運表現可望顯著成長。
自
自由时报GPU
Stabil73 Meldungen15 QuellenZuletzt aktualisiert: 5 sa önce
Neueste Meldungen
In Entwicklung
Business·6 sa önceKI-Zusammenfassung
字節跳動洽談採購天數智芯AI晶片，擬成其第三大國產GPU供應商
中國科技巨頭字節跳動正與上海AI晶片公司天數智芯洽談採購用於AI推理的晶片，擬成為其第三大國產GPU供應商。此舉顯示中國晶片製造商在政府推動科技自主及美國限制先進晶片出口的背景下，正努力提供外國AI晶片的替代方案。
自
自由时报
In Entwicklung
Technik·8 sa önceKI-Zusammenfassung
Robôs aprendem a "enxergar" com inteligência artificial e visão computacional
Robôs estão aprendendo a interpretar o mundo visual através da inteligência artificial e visão computacional, com aplicações em carros autônomos, indústria e medicina.
G
G1
In Entwicklung
Technik·13 sa önceKI-Zusammenfassung
AI伺服器進入「光銅並進」新架構，傳輸技術成關鍵
AI產業競爭轉向系統效率，高效率機櫃部署需GPU高速交換資料，光傳輸技術成焦點。AI資料中心迎「光銅並進」新架構，銅纜用於機櫃內短距離，光通訊用於跨機櫃長距離傳輸。貿聯-KY與佳必琪等廠商積極布局相關技術與產品。
自
自由时报
In Entwicklung
Technik·22 sa önceKI-Zusammenfassung
GPU DePIN : la promesse d'une IA décentralisée tient-elle enfin ?
Les réseaux GPU DePIN visent à concurrencer les géants du cloud en louant la puissance de calcul dispersée mondialement. Malgré une promesse vieille de dix ans, le secteur peine à capturer de la valeur, avec seulement Render montrant des signes de succès commercial durable.
J
Journal du Coin
In Entwicklung
Technik·1 g önceKI-Zusammenfassung
İletişim Başkanlığı'ndan Yerli ve Milli Yapay Zeka Ekosistemi Açıklaması
Cumhurbaşkanlığı İletişim Başkanlığı, güvenli, yerli ve kamuya açık bir yapay zeka ekosistemi inşa ediyor. Amaç, veriyi koruyan ve yapay zekayı kamu yararına kullanan güçlü bir altyapı oluşturmak.
M
Milliyet Son Dakika
In Entwicklung
Technik·2 g önceKI-Zusammenfassung
Fresh Tech Reviews: Foldable Phones, GPUs, Headphones, and More
New reviews cover foldable phones like the Motorola Razr Ultra and Honor Magic V6, an affordable AMD RX 9070 GRE GPU, Marshall's Milton ANC on-ear headphones, Logitech's Mobi Fold mouse, and initial impressions of the Rivian R2 SUV.
E
Engadget
In Entwicklung
Technik·2 g önceKI-Zusammenfassung
Microsoft, Yerel Yapay Zeka Özelliklerini Genişletiyor
Microsoft, yerel yapay zeka özelliklerini Copilot+ bilgisayarlarla sınırlı tutma politikasını gevşeterek, Nvidia RTX 30 serisi ve üzeri GPU'lara sahip Windows 11 cihazlara da genişletiyor. Bu güncelleme, geliştiricilerin dil modeli API'lerini NPU'suz sistemlerde de kullanabilmesini sağlıyor, ancak Recall ve Click to Do gibi bazı özellikler NPU'ya özel kalmaya devam ediyor.
S
ShiftDelete
In Entwicklung
Business·2 g önceKI-Zusammenfassung
MSI: Bellek ve GPU Kıtlığı 2026'ya Kadar Sürecek, CPU Arzı 3. Çeyrek'te İyileşecek
MSI, bellek ve GPU tedarik sıkıntılarının 2026'ya kadar devam edeceğini öngörüyor. Ancak, CPU arzının 3. çeyrekte iyileşmesi bekleniyor. Yapay zeka talebi ve bellek üreticilerindeki belirsizlikler fiyatları etkiliyor.
S
ShiftDelete
In Entwicklung
Business·2 g önceKI-Zusammenfassung
GPU Designer MetaX Plans Hong Kong Listing
Shanghai-based GPU designer MetaX announced plans for a Hong Kong listing, aiming to raise capital for R&D, expansion, and strategic investments. The proposed H-share offering, representing up to 5% of enlarged capital, follows its recent debut on Shanghai's Star Market.
S
SCMP Tech
NACHRICHT
4 g önceKI-Zusammenfassung
KB증권, SK텔레콤 목표주가 15만원으로 상향…AI 인프라 사업 기대
KB증권이 SK텔레콤의 AI 인프라 사업, 특히 GPUaaS와 AI 데이터센터 사업의 성장 가능성을 높이 평가하며 목표주가를 13만원에서 15만원으로 상향 조정했다. GPUaaS는 엔비디아와의 파트너십을 통해 추가 모멘텀을 확보했으며, AI 데이터센터 사업은 빠른 진척과 가격 프리미엄을 통해 실적 개선을 이끌 것으로 전망된다.
연
연합뉴스
In Entwicklung
Technik·5 g önceKI-Zusammenfassung
Xiaomi Breaks AI Inference Speed Record with MiMo-V2.5-Pro-UltraSpeed
Xiaomi achieves AI inference speed record with MiMo-V2.5-Pro-UltraSpeed, processing over 1,000 tokens per second on a single 8-GPU node, outperforming competitors with software innovations.
D
Decrypt
Technik
07.06.2026KI-Zusammenfassung
SpaceX、GoogleとのAIコンピューティング契約で月額1440億円の収益見込む
SpaceXはGoogleとの契約で、AIコンピューティング能力を提供し月額約1440億円の収益を目指す。約11万個のNVIDIA GPUを提供する2026年10月から2029年6月までの契約で、Googleは予想超の需要に対応するためのつなぎキャパシティを確保。
I
ITmedia
Technik
06.06.2026KI-Zusammenfassung
AMD Radeon RX 9070 GRE Review: A Shrinkflation GPU in a Tough Market
AMD's Radeon RX 9070 GRE launches at $549 with reduced specs (85% GPU cores, 75% memory, 66% memory bandwidth of RX 9070), feeling like 'shrinkflation'. It performs 10-20% slower than RX 9070 at 1440p, often slower than Nvidia's RTX 5070, especially with ray-tracing. While it matches some older high-end Nvidia cards, its 12GB RAM limits 4K performance, and it consumes the same power as the full RX 9070, making it less efficient.
A
Ars Technica
NACHRICHT
04.06.2026KI-Zusammenfassung
Губернатор Архангельской области и Hainan Yangpu NewNew Shipping обсудили "Арктический экспресс"
Губернатор Архангельской области Александр Цыбульский и председатель Hainan Yangpu NewNew Shipping Фан Юсинь обсудили развитие "Арктического экспресса" и проект международного порта Архангельск на ПМЭФ. Сотрудничество включает регулярные контейнерные перевозки между Россией и Китаем, сокращающие время доставки почти вдвое по сравнению с Суэцким каналом.
Р
РИА Новости
In Entwicklung
Technik·04.06.2026KI-Zusammenfassung
韓媒：樂金集團將採用1萬顆輝達GPU
韓國媒體報導，樂金集團將採購1萬顆輝達GPU用於AI模型訓練及人形機器人研發。輝達執行長黃仁勳將訪問韓國，與韓國主要企業領袖會面探討合作。
中
中央社
NACHRICHT
04.06.2026KI-Zusammenfassung
Huangpu Customs Implements "Cloud Issuance" for Hong Kong-Bound Poultry
Huangpu Customs has successfully implemented a "cloud issuance" system for animal health certificates, enabling the swift export of 10,000 chicks to Hong Kong. This digital process significantly reduces processing time and ensures the quality and safety of live poultry destined for the market.
中
中国新闻网
In Entwicklung
Technik·04.06.2026KI-Zusammenfassung
永擎攜手日本Data Section 戰略合作 搶攻AI基礎建設商機
伺服器廠永擎宣布與日本上市公司Data Section達成戰略合作，永擎將售出587台搭載輝達B200 GPU的AI伺服器，共計4696顆GPU，部署於泰國曼谷近郊AI資料中心，作為美國大型企業專屬算力平台，以滿足東南亞市場對高效能AI算力的需求。
自
自由时报
Technik
03.06.2026KI-Zusammenfassung
麗臺董事長盧崑山：看好下半年接單與業務發展維持正向成長
麗臺董事長盧崑山在任內最後一次股東會上表示，公司已從GPU代理轉型為實體AI與落地式AI解決方案市場，並成立新事業處切入工作站、AI伺服器與液冷應用。他看好下半年接單與業務發展，並預計由大訊董事長梁見發接任董事長。
自
自由时报
Technik
03.06.2026KI-Zusammenfassung
Lenovo Japan Unveils New ThinkPad and ThinkStation Workstations
Lenovo Japan announced new ThinkPad and ThinkStation workstations on June 2nd, including 7 models for 2026. The lineup features AMD Ryzen AI PRO 400 and Intel Core Ultra Series 3 CPUs, with options for NVIDIA RTX PRO GPUs and up to 96GB RAM. The ThinkStation P4 introduces liquid cooling.
I
ITmedia
In Entwicklung
Business·03.06.2026KI-Zusammenfassung
健策總座：ASIC客戶下半年出貨量將超GPU，看好兩片式均熱片趨勢
散熱股王健策總經理林錦隆表示，均熱片等產品需求正向，預計下半年起數家ASIC客戶出貨量將超越GPU。他強調，產業趨勢仍將走向兩片式均熱片，並看好Micro Channel Lid（微流道均熱片）及VC MC一體式散熱方案的未來發展，公司規劃2027年擴增台灣產能。
自
自由时报
In Entwicklung
Technik·02.06.2026KI-Zusammenfassung
Nvidia ra mắt chip RTX Spark, tích hợp CPU Arm Grace và GPU Blackwell RTX
Nvidia công bố chip RTX Spark tại Computex 2026, tích hợp CPU Arm Grace và GPU Blackwell RTX, hướng tới người sáng tạo nội dung, nhà phát triển AI và game thủ. Chip dự kiến ra mắt tháng 9 trên các thiết bị của Dell, Lenovo, Microsoft, HP, Asus, MSI.
V
VnExpress
In Entwicklung
Technik·02.06.2026KI-Zusammenfassung
Nvidia lança RTX Spark GPU para PCs com IA em meio a incertezas de demanda
Nvidia lança a GPU RTX Spark para PCs com IA, visando processamento local de tarefas de inteligência artificial. A aposta ocorre em um mercado com demanda incerta, apesar de a HP registrar alta nas vendas de PCs com IA.
G
G1
Technik
02.06.2026KI-Zusammenfassung
AMD Radeon RX 9070 GRE GPU Launching in Japan on June 2
AMD's new desktop GPU, the Radeon RX 9070 GRE, will be available in Japan from June 2nd. While pricing is unannounced, it's expected to be around $549 in the US, suggesting an accessible price point.
I
ITmedia