Huawei is a Chinese technology company active in telecom equipment, cloud computing and AI chips. Recent coverage has focused on its AI chip lineup: the Ascend 960DT was moved up to Q1 2027, prices for its most advanced data-center AI accelerators rose about 60% this summer, and it secured a 1.5 billion yuan loan for chip purchases. DeepSeek also released chip programming tools built with Huawei, and Qualcomm licensed its LogicFolding patents.
The South China Morning Post reports that Huawei and DeepSeek are narrowing the AI chip ecosystem gap with Nvidia, framing the trend as 'Jensen's nightmare.' The excerpt provides no specific performance benchmarks or market figures.
Huawei Technologies secured a 1.5 billion yuan ($2.25 billion) loan from Chinese banks to fund chip purchases for AI-related projects. The move reflects Huawei's increased fundraising amid soaring AI chip spending.
China is expanding its AI footprint in Latin America through investments in data centers, cloud platforms, and submarine cables, with Huawei, Alibaba, and ByteDance embedding themselves in regional ecosystems. Analysts warn of hard-to-reverse technological dependence and urge Latin American governments to diversify suppliers.
Qualcomm has agreed to license patents underpinning Huawei's LogicFolding chipmaking technique, seen as a win for Huawei's push into overseas AI markets.
A user built a local inference machine with two Huawei Atlas 300I Duo cards (96GB each, actually two 48GB chips per card). After vLLM, driver and operator optimizations, Qwen3.8 Flash-Next went from about 1 token/s to roughly 30 tok/s single-request and 61 tok/s at four-way concurrency, completing all 198 GPQA Diamond questions.
DeepSeek has publicly released software it built with Huawei to program AI chips, underscoring their close partnership aimed at replacing Nvidia technologies. The move could strengthen China's domestic AI chip ecosystem.
Huawei Cloud launched its AI Cluster Service, supporting over 40 days of stable cloud training, 10-minute fault recovery, and 20% higher token throughput, with commercial availability in China on September 30 and abroad November 30. It also highlighted an Agentic MaaS platform, the AgentArts agent platform serving 100+ enterprises, and the Industry AI Foundry.
Huawei moved up the launch of its next-generation AI chip series, with the Ascend 960DT now slated for Q1 2027 and a 960PR variant in Q3, aiming to replace Nvidia in China and compete globally.
Huawei executive Xu said Chinese AI is not yet powerful enough to pose 'frontier risks', implying current Chinese models remain behind the most advanced systems and that extreme-risk urgency is lower for now.
Huawei plans to unveil new AI technology this week aimed at challenging Nvidia's dominance in China and competing globally despite U.S. export controls, signaling a push by Chinese firms into high-performance AI chips.
Huawei forecasts that by 2035 billions of agents will dominate AI traffic, implying major implications for network, compute architecture and traffic management in the future.
UNESCO highlighted at its “Learning with new technologies and artificial intelligence” event that digital tech and AI are reshaping learning, urging education as a right and promoting fairness, inclusion, and skills training.
Huawei told some customers it raised prices of its most advanced data-center AI accelerators by about 60% this summer, reflecting demand far outpacing supply. The move signals notable price pressure on high-end AI chips and will affect cloud and compute procurement costs.
ASML convinced TSMC, Samsung and Intel to switch to larger photomasks, which should boost throughput of its newest EUV machines by about 40%. Meanwhile Huawei, via equipment maker Yuliangsheng and its suppliers, is pursuing a domestic strategy to reduce dependence on Dutch lithography technology.
That is everything