Development Hardware

The Dragonwing IQ-9075 EVK is a comprehensive evaluation platform to test the industrial-grade performance of the Dragonwing IQ-9075 processor for IoT solutions. The Dragonwing IQ-9075 EVK showcases the most common features of the Dragonwing IQ-9075 processor, which is optimized for industrial IoT and robotics applications involving heavy workloads in the most demanding environments like autonomous mobile robots (AMRs) and industrial machinery. The Dragonwing IQ-9075 EVK consists of a mainboard, supporting key interfaces, and optional expansion mezzanine boards to support additional features. It enables developers, OEMs, and ODMs to design and test their solutions powered by the industrial-grade Dragonwing IQ9075 processor integrated with standard interfaces and connectors, mezzanine expandability, and wireless connectivity. Test and prototype software development and hardware designs with less effort and cost to get your product to market faster.

The Dragonwing IQ-9075 Module is a highperformance, industrial-grade, AI-enabled module designed to simplify product design, streamline SoC integration, and shorten time to market. Built on a Dragonwing IQ9 Series processor for exceptional on-device AI performance, high-compute capabilities, built-in safety and reliability features, and design flexibility, this module is designed to meet the needs of the most demanding industrial-grade workloads in extreme environments. The processor supports Ubuntu and upstream Linux with Yocto support, and the MCU subsystem runs RTOS for easy deployment. Housed in a 1573 BGA package, the Dragonwing IQ-9075 Module features 36 GB of ECC-enabled LPDDR5 memory and supports a wide ambient operating temperature range from -40 to +85 °C. This pin-to-pin-compatible module is optimized for long lifecycle deployments in Industry 4.0 environments.

The Open-Q Qualcomm Dragonwing™ 9075IQ System-on-Module (SOM) solutions offer scalable, power-efficient, and robust computing capabilities for autonomous devices and next-generation Industry 4.0 designs. These advanced AI-powered SOMs enable a range of critical functionalities, including robust safety features for autonomous mobile robots with enhanced device robustness through fault-tolerant ECC memory support and a dedicated real-time subsystem. The new SOM series facilitates the deployment of interactive industrial edge AI systems by integrating Large Language Model (LLM) support, achieving a generation rate of up to 22 tokens per second when running the LIama2 7B parameter model.

MeiG SNM983 series module is based on the Qualcomm Kryo Gen 6 CPU with a maximum frequency of 2.36 GHz and a 5nm NON-POP process. With Qualcomm Adreno VPU 765, it supports UHD170 video enconding/UHD275 decoding, H.264/H.265. Adreno 663 GPU supports safe GPU compute, up to 840 MHz, OpenGL ES 3.2, Vulkan 1.2, and OpenCL 2.2. Integrated with Qualcomm neural network processing unit Qualcomm® Hexagon™ Tensor Processor, it supports HVX, HMX and models like ONNX, PyTorch, TensorFlow, CAFFE, and TensorFlow Lite, its computing power is up to 100 Tops. SNM983 series module integrates rich functional interfaces, including DSI, DisplayPort, CSI, PCIe, SGMII, RGMII, UFS, USB, I2C, UART, and SPI interface, etc. Products can be widely used in AI BOX, video bar, VR camera, intelligent robots, intelligent information gathering equipment, live streaming terminals and other products.

The AOM-6741 is a SMARC 2.2 full-size computer-on-module powered by the Qualcomm Dragonwing IQ-9075 processor. It features octa-core Kryo Gen6 CPUs and a 5th-generation VPU, delivering up to 100 TOPS of edge AI performance. The module supports 4 × 4-lane MIPI-CSI (D-PHY) and 4 × DisplayPort, making it ideal for vision intelligence applications. Ubuntu Pro is supported, providing up to 10 years of Expanded Security Maintenance for Ubuntu Classic LTS.

This collaboration centers on the EXMP-Q911 COM-HPC Mini module, powered by Qualcomm Dragonwing™ IQ-9075 SoC featuring an 8-core Qualcomm® Kryo™ Gen 6 CPU and a Qualcomm® Adreno™ 663 GPU, enabling 100 TOPS dense or up to 200 TOPS sparse AI compute. Under defined 30W test scenarios, the EXMP-Q911 achieves up to 10× higher AI inference FPS compared to similar modules when running the YOLOv10n object detection model across 10 concurrent video channels. Designed for demanding edge workloads, the EXMP-Q911 integrates 36GB LPDDR5X memory and 128GB UFS 3.1 storage, paired with a rich set of high-speed interfaces, including PCIe Gen4, USB 3.2, dual 2.5GbE LAN, dual DP 1.2, MIPI CSI-2, CAN FD, and more, providing exceptional connectivity and performance density for compact edge AI systems.

The Innodisk EXEC-Q911 is a complete starter kit featuring a COM-HPC Mini Module and carrier board for immediate development and evaluation. The COM-HPC Mini Module leverages Qualcomm’s latest innovation—the Dragonwing™ IQ-9075 SoC—to deliver high AI computing performance with low power consumption. Combined with Innodisk’s comprehensive customization services and self-developed software toolkit, it provides customers with a fast, time-to-market solution to stay ahead in the edge AI era. Innodisk leverages this innovation to develop the EXEC-Q911 starter kit, integrating a COM-HPC Mini Module and carrier board to deliver ready-to-use, high-performance, reliable, and scalable on-device AI computing with low power consumption at the edge.