With growing demand for edge AI and ultra-HD capabilities, the Forlinx Embedded FET3572-C SoM stands out with its 4 TOPS NPU, 8K codec performance, and balanced power efficiency.
AIoT is driving intelligent edge applications to grow at an exponential pace. From real-time recognition in smart security systems to high-definition rendering on commercial displays, from edge computing workloads to intelligent interactions in portable devices, the market is demanding significantly more AI compute power and multimedia capability from system-on-modules (SoMs).
The key requirements are strong edge AI inference, ultra-HD video codec support, and a practical balance between performance and power efficiency.
In response, Forlinx Embedded, a strategic partner of Rockchip, has introduced the FET3572-C SoM, powered by the RK3572 processor.
Equipped with a hardware-accelerated 4 TOPS NPU and an octa-core heterogeneous architecture backed by a comprehensive ecosystem, this SoM delivers a versatile compute platform for edge AI scenarios and sets the stage for embedded intelligence innovations.
4 TOPS NPU for Faster, More Efficient Edge AI Inference
The FET3572-C includes a dedicated NPU delivering 4 TOPS (INT8) compute power, addressing the core bottlenecks of AI performance and inference speed at the edge. It supports INT4/INT8/INT16 mixed precision and works with frameworks like TensorFlow and PyTorch, making it straightforward to deploy diverse edge AI workloads.
Whether it’s facial recognition in smart security, anomaly detection in industrial environments, or intelligent noise reduction in consumer electronics, the FET3572-C enables real-time responses with its robust compute capability.
By removing dependence on cloud computing, it sharply reduces data transmission costs and latency, delivering true local intelligence for edge devices.
Performance and Power Efficiency in Balance
Powerful AI and multimedia features need a solid hardware foundation. The FET3572-C is built on an octa-core heterogeneous architecture (2×Cortex?A73 + 6×Cortex?A53) using an advanced 8nm process. This design delivers more than 100% higher performance than previous mid-range platforms while cutting typical power consumption by over 50%.
The dual A73 cores handle demanding AI inference and ultra-HD video processing, while the six A53 cores manage lighter tasks ensuring high performance when needed and optimal efficiency during low-load operations.
Antutu v10 scores exceed 310,000, and standby power stays below 10mW, enabling support for heavy workloads like the 4 TOPS NPU and 8K codec while meeting the low-power needs of portable devices.
8K Codec for True Ultra-HD Experience
The SoM includes a high-performance multimedia unit that supports 8K decoding and 4K encoding across mainstream and open-source formats.
It integrates a 12 MP ISP for accurate color reproduction and supports 5 camera inputs, ideal for multi-channel capture and monitoring.
With dual-screen independent display (4K@60fps + 2K@60fps), it’s well-suited for commercial displays, POS systems, digital signage, and more. The efficient decoding architecture reduces system bandwidth and power usage delivering exceptional 8K quality without compromising stability or energy efficiency.
Adaptability Across Applications
Beyond AI and multimedia, the board offers extensive connectivity:
- High-speed interfaces: PCIe 2.1, dual Gigabit Ethernet
- Industrial buses: CAN-FD, I2C
- Native support for LPDDR5/5X memory
It adapts seamlessly to smart security, commercial displays, industrial control, edge computing, and vehicle-mounted systems enabling everything from HD surveillance with real-time AI analysis to local data processing and intelligent interaction.
Comprehensive Ecosystem and Support
To accelerate development, the board supports both Linux and Android and comes with complete resources like the driver source code, development manuals, and technical examples.
Backed by Forlinx Embedded’s expertise, customers receive end-to-end support from solution design and prototyping through mass production helping shorten R&D cycles and accelerate time-to-market.
The All-in-One Edge AI Platform
With growing demand for edge AI and ultra-HD capabilities, the Forlinx Embedded FET3572-C SoM stands out with its 4 TOPS NPU, 8K codec performance, and balanced power efficiency.
It delivers an efficient, stable, and cost-effective core solution for upgrading intelligent devices across industries.
The Forlinx Embedded FET3572?C SoM is now open for pre-orders.
Developers and manufacturers are invited to explore collaboration let’s build more competitive intelligent products and drive the edge AI industry forward.
