Nvidia worker among 9 charged in Taiwan chip smuggling case
TAIPEI: Nine people in Taiwan, including an Nvidia worker, were charged on Monday (Aug 24) over the illegal export of artificial intelligence servers to China, Taiwanese prosecutor...
#AI Hardware
5346 articles tagged with AI Hardware
Exynos 2700, il chip di Samsung punta a battere Snapdragon 8 Elite Gen 6: cosa sappiamo dai nuovi leak Il futuro chipset...
Exynos 2700, il chip di Samsung punta a battere Snapdragon 8 Elite Gen 6: cosa sappiamo dai nuovi leak Il futuro chipset di punta di Samsung, l'Exynos 2700, continua a far parlare ...
Xiaomi Unveils New Xring Chip as It Seeks Greater Control Over Smartphone Technology
Xiaomi has unveiled a new generation of its proprietary smartphone processor as the Chinese electronics giant steps up efforts to develop key technologies in-house and reduce its d...
Wine 11.16 brings VA-API hardware video decoding, improved ARM64 support, and 35 bug fixes
Wine 11.16 is now available for Linux, macOS, and BSD, reinforcing its role as a compatibility layer for running Windows applications and games on POSIX-compliant systems. This rel...
Xiaomi launches new Xring chip, deepening push into in-house processors
Xiaomi unveils its Xring O3 smartphone processor, set for TSMC’s 3nm process, as it expands in-house chip development and reduces reliance on Qualcomm.
Gigant chipów prowadzi rozmowy o inwestycji, która wywinduje wartość Perplexity AI o ponad 50 proc. w niespełna rok. Fir...
Gigant chipów prowadzi rozmowy o inwestycji, która wywinduje wartość Perplexity AI o ponad 50 proc. w niespełna rok. Firma stawia na agentów AI i twardo deklaruje datę debiutu na g...
Xiaomi develops its own AI mobile chip in challenge to Qualcomm, MediaTek
The Xring O3 system-on-a-chip advances Xiaomi's in-house semiconductor design efforts with more power-efficient image processing and better artificial intelligence capabilities
🔌 NVIDIA ALZA I PREZZI DEI CHIP AII clienti più grandi sono stati avvisati: i server con chip Vera Rubin e Grace Blackwe...
🔌 NVIDIA ALZA I PREZZI DEI CHIP AII clienti più grandi sono stati avvisati: i server con chip Vera Rubin e Grace Blackwell costeranno +15% dal 2027. La corsa all'IA fa lievitare i ...
Compare product engineering companies by technical depth, not size. Strong partners combine hardware, software, electron...
Compare product engineering companies by technical depth, not size. Strong partners combine hardware, software, electronics, cloud, #AI testing, and domain expertise to deliver rel...
Stop Comparing GPU Clouds Only by $/Hour
Every GPU cloud comparison I see is a table of dollars per hour. Provider A is $2.10 an hour for an...
The top GPU neoclouds of 2026 have been ranked: CoreWeave commands a premium as the only Platinum-rated provider, while ...
The top GPU neoclouds of 2026 have been ranked: CoreWeave commands a premium as the only Platinum-rated provider, while Nebius posts the lowest H100 rate and the only published B30...
Amazon hikes Echo, Kindle, Fire TV prices as memory chip costs surge
Amazon has increased prices across several hardware products, including Echo, Kindle, Fire TV and eero devices, as rising memory and storage costs pressure the consumer electronics...
8GBのGPUで35Bモデルが高速動作！MoE特化の実行環境「FreeToken」https://pc.watch.impress.co.jp/docs/news/2134988.html#impress #市場 #AI #その他
8GBのGPUで35Bモデルが高速動作！MoE特化の実行環境「FreeToken」https://pc.watch.impress.co.jp/docs/news/2134988.html#impress #市場 #AI #その他
GPU undervolting during CNN training introduces stochastic perturbations that improve adversarial robustness and lower p...
GPU undervolting during CNN training introduces stochastic perturbations that improve adversarial robustness and lower power.Source: arXiv cs.LGhttps://arxiv.org/abs/2608.20572#Mac...
Stop Measuring GPU Speed! Why Agentic Workloads Require New Performance Metrics
Traditional metrics like GPU speed are failing. Discover why enterprise data centers must transition to measuring 'delivered agents per rack' to accurately assess performance for c...
Even with massive profit margins, NVIDIA can’t absorb the skyrocketing costs of high-bandwidth memory. The chip giant is...
Even with massive profit margins, NVIDIA can’t absorb the skyrocketing costs of high-bandwidth memory. The chip giant is raising AI server prices by over 15% starting early next ye...
Startup z ZEA rzuca wyzwanie gigantom chmury, oferując firmom dostęp do mocy GPU z gwarancją pełnej izolacji danych. Dzi...
Startup z ZEA rzuca wyzwanie gigantom chmury, oferując firmom dostęp do mocy GPU z gwarancją pełnej izolacji danych. Dzięki technologii Intel TDX klienci mogą samodzielnie weryfiko...
iPhone 18 Pro’s 2nm Chip: What It Could Mean for Performance and Battery Life
iPhone 18 Pro with A20 Pro could redefine Apple iPhone performance, while iPhone 18 Pro Max may bring better speed, efficiency and battery life.
🚀🤖 In a thrilling tale #of #nostalgia and #tech #jargon, #AI #chip architects reminisce about the good ol' days when CPU...
🚀🤖 In a thrilling tale #of #nostalgia and #tech #jargon, #AI #chip architects reminisce about the good ol' days when CPUs actually got faster 📈, only to now bet the farm on fancy #...
AI Chip Architectureshttps://www.jepeake.com/ai-chip-architecturesComments: https://news.ycombinator.com/item?id=4940565...
AI Chip Architectureshttps://www.jepeake.com/ai-chip-architecturesComments: https://news.ycombinator.com/item?id=49405657#HackerNews #AI #Chip #Architectures #technology #innovatio...
AI Chip Architectureshttps://www.jepeake.com/ai-chip-architectures#ai
AI Chip Architectureshttps://www.jepeake.com/ai-chip-architectures#ai
Etched Sohu vs. Nvidia: Transformer ASIC vs. GPU (2026) – Spheron Bloghttps://www.spheron.network/blog/etched-ai-sohu-vs...
Etched Sohu vs. Nvidia: Transformer ASIC vs. GPU (2026) – Spheron Bloghttps://www.spheron.network/blog/etched-ai-sohu-vs-nvidia-transformer-asic-inference/#ai #nvidia
My CUDA/GPU Journey: From "What Even Is a GPU?" to Actually Fascinated
A few months ago, if you'd asked me what a GPU actually does, I would've mumbled something about...
FreeToken makes it possible to run 753B parameter models like GLM-5.2 on a single GPU, opening new possibilities for edg...
FreeToken makes it possible to run 753B parameter models like GLM-5.2 on a single GPU, opening new possibilities for edge AI deployment.https://www.marktechpost.com/2026/08/23/meet...