AI端侧应用

Search documents
艾为电子(688798.SH)发布2024年度业绩,归母净利润2.55亿元,同比增长399.68%
智通财经网· 2025-04-09 12:49
Core Viewpoint - The company reported significant growth in revenue and net profit for the fiscal year 2024, driven by strategic product development and market expansion in high-performance sectors [1][2]. Financial Performance - The company achieved a revenue of 2.933 billion RMB, representing a year-on-year increase of 15.88% [1]. - The net profit attributable to shareholders reached 255 million RMB, showing a remarkable year-on-year growth of 399.68% [1][2]. - The basic earnings per share (EPS) stood at 1.1 RMB, with diluted EPS also reflecting a 400% increase year-on-year [2]. Product and Market Strategy - The company focused on high-performance mixed-signal, power management, and signal chain products, providing comprehensive solutions across design, algorithms, applications, and processes [1]. - The company expanded its product matrix and launched new products, capitalizing on the recovery of the electronic information industry and the growing demand for AI applications [1]. - The company achieved a historical high in shipment volume, exceeding 6 billion units, contributing to the revenue growth [1]. Operational Efficiency - The increase in net profit was attributed to enhanced revenue scale and efficiency, leading to improved profit margins [2]. - The company saw a continuous rise in market share for new products and sectors, resulting in a higher gross margin compared to the previous year [2]. - Ongoing management reforms and digital transformation in research and development have improved efficiency, further solidifying profitability [2].
营收大幅增长81%-95%,佰维存储前瞻性布局AI端侧应用迎来大收获
市值风云· 2025-01-23 11:39
先进封测是核心差异化能力。 | 作者 | | 小鑫 | | --- | --- | --- | | 编辑 | | 小白 | 在刚刚过去的CES 2025展会上,AI眼镜和AI智能玩具成为一大亮点。包括Halliday、Xreal、Rokid、 雷鸟创新等厂商一共带来了近50款AI(AR)眼镜,其中中国厂家撑起了半边天。 (图片来源:网络) Meta作为AR/XR领域的巨头之一,选择在Wynn酒店展示了Ray-Ban Meta AI眼镜系列以及Meta Quest 3、3S。其中,Ray-Ban Meta AI眼镜销量已经突破了200万台,其在人机交互等多个领域具有领先优 势,是很多厂商追赶的目标。 这一轮AI眼镜和智能硬件爆发的核心在于大模型能力在端侧的应用。目前市场主流的大模型参数都 在千亿以上,要想实现端侧AI,有三个步骤非常重要,分别是模型压缩、模型适配、人机交互。 模型压缩的目的在于降低对硬件的要求,这一步通常是大模型厂商在做,更多是软件层面的,比如谷 歌推出的Gemini Nano,Meta推出的MobileLLM。 模型适配包括存储的提升、芯片的适配。大模型对于存储的要求有目共睹,之前已经带火了H ...