Developer Kits with Intel® Core™ Ultra Processors (Series 3)
Preinstalled Software
Get the most out of your hardware performance with the Intel® Distribution of OpenVINO™ toolkit, Ubuntu* Desktop LTS, and libraries.
Pretrained Models for Acceleration
Choose from a variety of optimized detection and recognition models for developing deep learning applications.
Training Extensions for Deep Learning
Modify, customize, train, and extend computer vision models for deep learning and inference optimization.
Overview
Intel® Core™ Ultra processors can scale performance and take AI acceleration to a new level to supercharge edge computing in the era of AI. From digital safety in smart cities to AI automation on factory floors. From immersive interactivity in retail and entertainment to AI-assisted workflows in smart hospitals.
Enable edge AI and graphics in more places—wherever your business needs to be.
- Intel® Core™ Ultra processors offer a comprehensive suite of integrated compute engines tailored for AI, including the CPU, GPU, and NPU.
- Empower next-generation graphics and media workloads at the edge with enhanced built-in Intel® Arc™ GPUs.
- Enhance your edge device designs with greater flexibility and efficiency by leveraging a 12–65W TDP range. Available comprehensive long-term software support helps maximize deployment value and extend device longevity.
Who Needs This Product
System integrators, independent software vendors (ISV), and IoT developers who create vision-based inferencing applications at the edge that demand:
- High-performance AI inferencing algorithms across Intel® Vision Products and platforms.
- Enjoy performance for AI-driven smart city use cases. Support for multiple cameras per device can help enable cost-effectiveness, manageability, and scalability in settings that need extensive video monitoring capabilities.
- Enhance productivity and safety on shop floors and support advanced Industry 4.0 use cases—including AI vision deployments—while helping to consolidate workloads on easy-to-manage systems in hard-to-reach places. Benefit from cost-efficient platform consolidation, a high number of video streams per device, and detail-rich interfaces.
To determine the features and capabilities implemented by our ecosystem collaborators, see Hardware.
Reference Implementations
Prebuilt and validated reference implementations are available for developers to test and deploy industrial, retail, and healthcare applications enabled for computer vision.
Your experience may vary depending on the configuration of your developer kit. For details, see the target system requirements in the reference implementation.
Hardware
Prevalidated developer kits.
AAEON* UP Xtreme ARL
- Upgraded connectivity includes 2× USB4 (Type‑C), 2× USB 3.2 Gen2, dual 2.5GbE, 40‑pin GPIO, and 2× COM ports for industrial integration
- Display subsystem supports 4× 8K output via HDMI 2.1 and DP 2.1 (USB‑C), enabling multi‑camera, analytics, and monitoring workloads
- Offers dual M.2 2280 M‑Key storage slots + M.2 2230 E‑Key for wireless modules Memory support up to 128GB DDR5‑7200 across two SODIMMs
- Industrial-Grade Design & Reliability : wide voltage input range of 19V to 36V DC
AAEON* CEXD-INTRBL
- All‑in‑one development platform tailored for sophisticated robotics development
- Integrates eight GMSL2 camera inputs to enable synchronized, low‑latency multi‑camera vision pipelines
- Includes industrial communication interfaces for sensors, actuators, and real‑time control
- Designed to harness the advanced AI and I/O capabilities for robotics
ASRock Industrial* NUC(S) 300 BOX Series
- Supports up to 128GB DDR5‑7200 and dual M.2 (Gen5 x4 + Gen4 x4) with Intel® VMD RAID 0/1
- Quad 8K display capability via HDMI 2.1 + DP 2.1/1.4a for immersive visualization and multitasking
- Advanced connectivity with USB4/Thunderbolt™ 4, USB 3.2 Gen2, and dual 2.5 GbE LAN (model‑dependent)
- Wi‑Fi 7 + Bluetooth 5.4 for ultra‑fast wireless throughput and low‑latency device links
Innodisk AXMB-D160
- Features Intel Core Ultra X7 Processor 358H, delivering 180 TOPS for edge AI computing
- Equipped with Intel Core Ultra Series 3 platform-certified components: - Innodisk 4TG2-P series 512GB PCIe Gen 4 x4 M.2 SSD
- Mini-ITX form factor (170 x 170mm
ASUS NUC 16 Pro Mini PC
- AI‑accelerated compact platform powered by Intel® Core™ Ultra Series 3 with integrated Arc™ GPU and NPU, delivering up to ~180 TOPS for real‑time AI workloads and advanced analytics [
- Hyper‑connected I/O includes dual 2.5GbE LAN, Wi‑Fi 7, Bluetooth 6.0, Thunderbolt 4, and multiple USB ports, enabling low‑latency networking and flexible device integration [
- High‑performance memory & storage flexibility with up to 96GB LPDDR5x or 128GB DDR5 support and dual M.2 (PCIe Gen5 + Gen4) SSD slots for scalable performance and storage expansion
- Compact enterprise design & reliability featuring a 5×4 mini‑PC chassis, dual‑fan cooling, tool‑less upgrade architecture, integrated security (fTPM / vPro), and multi‑display support for business deployments
Advantech ARK-2252
- AI‑ready edge platform powered by Intel® Core™ Ultra Series 3 (Panther Lake), supporting up to 128GB DDR5‑7200 memory and M.2 PCIe Gen5 NVMe storage for high‑performance compute and analytics
- Upgraded industrial connectivity includes up to 4× 2.5GbE LAN, 6× COM, 6× USB, dual CANBus, and programmable DIO, enabling seamless integration with sensors, PLCs, cameras, and legacy devices
- Flexible expansion architecture with multiple M.2 (M‑Key/B‑Key/E‑Key) slots supporting storage, 5G/Wi‑Fi, and Advantech iDoor / PoE / vision modules for modular system customization
- Industrial‑grade fanless design & reliability featuring −20°C to 60°C operation, 12–24V wide power input, and triple 4K display output via HDMI and USB‑C for rugged edge AI deployments