据快科技报道，知名分析师郭明錤近日预测，全球图形处理器巨头英伟达的下一代人工智能（AI）芯片——R系列，具体型号为R100，预计将在2025年第四季度开始量产。这款备受期待的芯片将采用台积电先进的N3制程工艺和CoWoS-L封装技术，与英伟达当前的B100芯片保持一致。值得注意的是，R100的芯片设计规模扩大，采用约4x reticle设计，相比B100的3.3x reticle设计有所提升。
当前，AI服务器的高能耗问题已经成为云服务提供商（CSP）和超大规模数据中心（Hyperscale）在采购和建设数据中心时面临的关键挑战。英伟达显然已经洞察到这一市场趋势，因此在R系列芯片及其系统方案的设计中，不仅致力于提升AI计算性能，还特别强调了能效比的优化。这表明，R100将致力于提供更高效能、更低功耗的AI解决方案，以满足行业对于绿色、节能AI技术的迫切需求。
郭明錤还预测，基于R100芯片的系统和机柜方案预计将在2026年上半年实现量产，这将进一步推动AI计算领域的发展，为数据中心和云计算行业带来更为高效且节能的计算能力。英伟达的这一举措，无疑将对全球AI芯片市场产生深远影响，引领行业向更加可持续的方向发展。
英语如下：
News Title: “NVIDIA’s Next-Gen AI Chip R100: Mass Production to Commence by End of 2025, Targeting High Performance and Low Power Consumption Market”
Keywords: NVIDIA, AI chip, power optimization
News Content: According to Kuaitech, renowned analyst Ming-Chi Kuo recently predicted that NVIDIA, the global leader in graphics processing units (GPUs), is set to commence mass production of its next-generation artificial intelligence (AI) chip, the R series, specifically the R100, in the fourth quarter of 2025. This much-anticipated chip will utilize TSMC’s advanced N3 manufacturing process and CoWoS-L packaging technology, maintaining parity with NVIDIA’s current B100 chip. Notably, the R100’s chip design scale expands, adopting a roughly 4x reticle design, an upgrade from the B100’s 3.3x reticle design.
Currently, the high power consumption of AI servers has emerged as a key challenge for cloud service providers (CSPs) and hyperscale data centers (Hyperscale) when acquiring and constructing data centers.显然, NVIDIA has identified this market trend, hence in the design of the R series chips and its system solutions, it not only aims to enhance AI computational performance but also places a particular emphasis on power efficiency optimization. This suggests that the R100 will strive to offer a more efficient and power-saving AI solution, catering to the industry’s pressing need for green and energy-efficient AI technologies.
Kuo also forecasts that systems and rack solutions based on the R100 chip are expected to enter mass production in the first half of 2026, further propelling the development of AI computing and bringing more efficient and power-saving computational capabilities to data centers and the cloud computing industry. NVIDIA’s move is poised to have a profound impact on the global AI chip market, steering the industry towards a more sustainable trajectory.
【来源】https://news.mydrivers.com/1/978/978569.htm