The trend to integrate Generated Artificial Intelligence (GEN AI) applications with Edge AI is gaining momentum. More EDGE AI systems offer intelligent applications with multi-layer user interaction, stimulating opportunities for EDGE AI's explosive growth. For example, CES 2025 saw the announcement of new products such as AI Toys, AI Pets, AI Robots, and AI Robot Dogs. New applications are increasingly using compact, energy-efficient smart devices combined with AI inference and ample computing power.
Edge AI now calls for higher performance, privacy, response speed and efficiency. However, the industry is actively tackling this tough challenge due to the limitations of miniaturized devices in computing power, memory capacity and battery size. Thanks to the rapid iteration of cutting-edge semiconductors and AI technology, microcontroller (MCU) semiconductor solutions supporting neural processing units (NPUs) are rapidly emerging in the market. We aim to meet these requirements with the aim of quickly meeting AI MCU products that combine security, energy savings and computing power.
Alif launches new ensemble MCU series E4, E6 and E8 products
Alif Semiconductor, a global leader in AI MCU technology, announced the latest addition to its new generation of ensemble MCU families on August 12, 2025. Three new models, E4, E6 and E8, are designed to handle the workloads required for generated AI inference on EDGE AI devices. Targeting battery-powered use cases, these products allow for rapid execution of generated AI applications on compact devices such as wearables. They have ultra-low power consumption and core information security features. By eliminating the need to frequently charge and reduce the user experience, Alif's new generation ensemble E4/E6/E8 MCU solution establishes a major position in the EDGE AI market.
From a product specification perspective, the E4, E6, and E8 products feature functional configurations ranging from dual-core, triple-core to quad-core microprocessors. For example, the E4 chip has two cortical-M55 MCU cores, each paired with an EthoS-U55 NPU, and also an EthOS-U85 NPU that the Cortex-M55 can access for generation AI tasks. This, coupled with a flexible built-in memory configuration, including up to 5.5MB of MRAM and up to 9.75MB of SRAM, greatly facilitates the execution of a variety of AI models.
The E6 chip adds a cortex-A32 core to create a tri-core microprocessor to address applications requiring higher computing power for AI model inference. The highest performance E8 chip utilizes a quad-core microprocessor configuration and combines AFOREMentioned Configuration with a Cortex-A32 core to meet the demands of highly complex generation AI inference. Through this new product combination, Alif is optimistic about the new business opportunities for generation AI applications in subsequent Consumer Electronics products, and is poised to take advantage of the diverse Edge AI applications and thriving business opportunities.
The ETHOS-U85 NPU is configured to run a transformer-based neural network system with high efficiency
Alif's new ensemble MCU is equipped with an ARM ETHOS-U85 NPU, making it the industry's first MCU product to be released with this NPU. It supports trans-architecture-based generated AI neural networks and provides ultra-low energy efficiency. Current customer test data shows some AI toys that can automatically generate and compile story content tailored to a child's age, emotion, and personalization. These AI toys also have voice control for interactive interaction with children, representing new applications that combine educational and entertainment values, creating optimistic expectations for the smart toy market and the broader interactive smart devices.
Accelerated image processing enables object detection in 2 ms and image classification in less than 8 ms
In terms of image processing performance, the new chip currently supports up to two MIPI-CSI image sensors and includes a fully hardware-accelerated Image Signal Processor (ISP) channel. Thanks to the design of a high-performance memory subsystem, high-speed video images are captured up to 60 frames (FPS) at a resolution of 2 megapixels. In particular, the chip is equipped with ultra-energy-efficient and extremely flexible MRAM memory, which can reduce AI inference speeds to less than 1 millisecond.
Given that the generation AI plays a key role in optimizing intelligent robotics, not only does it improve the accuracy and adaptability of autonomous robots, but the new E4/E6/E8 products in the new generation ensemble MCU series can enhance real-time data collection and analysis capabilities for applications such as health monitoring and withstandable devices. This enables OEM/ODM electronic product development teams to launch a new generation of user experiences that create great interactive experiences with generated AI. This can be extended to include solutions in areas such as healthcare, smart healthcare and auxiliary diagnostics, smart transport, smart toys, smart homes, smart cities, and smart robots to drive innovation in next-generation AI products.
For more information, please visit the Alif Semiconductor website.
Ensemble Family
Article edited by Jack Wu
