智通财经APP获悉，东吴证券发布研报称，2026年AI算力产业链预计将迎来业绩放量。云端算力方面， 国产GPU进入业绩兑现期，产业链协同至关重要。端侧算力则受益于AI终端创新与3D DRAM放量，多 场景加速落地。同时，先进制程晶圆代工扩产、HVDC电源架构升级以及光铜互联需求增长，共同驱动 产业链各环节技术升级与价值提升。 东吴证券主要观点如下： 云端算力：迎接国产算力产业链上下游共振带动业绩放量 3DDRAM：端侧AI存储26年为放量元年，产业趋势逐渐确立 26年AI硬件落地带来存力需求的快速提升，高带宽/低成本的3DDRAM有望在多领域放量。该行认为机 器人/AIOT/汽车等领域对本地大模型的部署离不开3DDRAM存储的支持，其为各端侧应用从"能 用"到"好用"的关键硬件革新。多款NPU的流片发布亦为3DDRAM提供丰富适配场景。此外，手机/云端 推理等场景亦将逐步导入，成为26H2及27年关键场景，应用领域持续拓宽。相关公司兆易创新、北京 君正等。 端侧AI模型：架构优化突破物理瓶颈，利益分润决定生态版图 2026年国产算力芯片龙头有望进入业绩兑现期，看好国产GPU受益于先进制程扩产带来的产能释放。考 ...

来源：市场资讯 来源：半导体前线 机构认为，南亚科第3季以来受惠客户需求急速升温，DDR4价格一路飙涨，尤其三大原厂纷纷将产能 转为生产DDR5、HBM4等，不易回头生产DDR4，如今DDR4仍占有整体DRAM市场约20%，在庞大供 需缺口带动下，DDR4呈现每月报价调涨态势，助攻南亚科营运。 机构指出，AI服务器需求急速飙升，加上三星、SK海力士、美光三大原厂积极调整产能配置，优先支 持DDR5与HBM（HBM3E/HBM4）生产，使传统DDR4供给快速收敛、供需更形紧俏。 各大内存原厂在2025~2026年主力皆投向新世代产品，导致DDR4产能缩减幅度超过预期。 根据南亚科公告，10月营收79.08亿元新台币，较去年同期成长262.37%；税前净利25.02亿元新台币、 年增781.74%；归属母公司业主净利20.39亿元新台币，年增幅更达1006.38%。 全球DRAM价格自8~9月急速拉升，带动南亚科第三季营收187.79亿元新台币，年增130.9%；税前净利 20.23亿元新台币，较去年同期大幅成长210.67%；归属母公司净利15.63亿元新台币，年增205.16%，营 运已明显脱离谷底、迈入回升 ...

Group 1 - The core viewpoint of the report is that the transition from 2D to 3D architecture in DRAM is essential due to the challenges faced in further miniaturization of DRAM processes, with hybrid bonding technology representing the future path for 3D DRAM [1] - The report highlights that the current bottleneck in AI edge inference speed is memory bandwidth rather than computing power, and that 3D DRAM can significantly enhance transmission efficiency [1][3] - The use of NPU as a co-processor combined with 3D DRAM is likely to be the next trend in edge technology, with companies like Zhaoyi Innovation (兆易创新) being recommended for investment [1][2] Group 2 - AI applications are diversifying rather than conforming to a unified model, with new hardware technologies being developed to support this trend, particularly through the use of smaller models that outperform larger ones [2] - The report indicates that the memory bandwidth limitation is significantly greater than the computational limitation, as demonstrated by the Snapdragon 8 GEN3 example, where memory bandwidth constraints are evident [3] - Major players in the industry, including Zhaoyi Innovation and its subsidiaries, as well as Taiwanese storage IDM Winbond and Qualcomm, are focusing on the 3D DRAM and NPU solution, indicating a clear technological trend [3]

Summary of Qualcomm NPU + 3D DRAM Expert Exchange Takeaways Industry and Company Involved - The discussion revolves around **Qualcomm** and its collaboration with domestic suppliers in the **semiconductor** industry, specifically focusing on **NPU (Neural Processing Unit)** and **3D DRAM** technologies. Core Points and Arguments 1. **Collaboration Timeline**: Qualcomm officially initiated the project in Q1 of this year, completing validation with domestic suppliers. The expectation is to return the chips by the end of June for packaging, testing, yield, and functionality evaluation. Samples are planned to be sent to smartphone manufacturers between February and March next year, with a market launch anticipated in September to October [1]. 2. **Application Layer Upgrade**: The Qualcomm Snapdragon 8 Elite flagship model is equipped with LPDDR5x, offering a bandwidth of 85GB/s. After upgrading to 3D DRAM, the bandwidth is expected to exceed 1TB. Currently, the products under validation have a bandwidth of 256GB/s, representing a threefold increase in speed. The total cost increase is approximately $60, making it suitable for high-end models that utilize computational photography and AI [2][3]. 3. **WoW Solution Advantages**: Compared to Winbond's CUBE, the WoW solution utilizes hybrid bonding, which eliminates micro bumps, resulting in a thinner stacking height. This provides advantages in memory bandwidth and power consumption. Additionally, the yield rate for stacks exceeding four layers is superior to that of CUBE [4]. Other Important but Possibly Overlooked Content - The discussion highlights the competitive edge of Qualcomm's technology in the semiconductor market, particularly in high-performance applications, which may present significant investment opportunities in the sector as demand for advanced mobile computing continues to grow [5].