Tria Technologies Brings On-Board AI Performance to new Levels with Innovative Compact COM Express Module

­Tria Technologies, an Avnet company specialising in embedded compute solutions, has launched the Tria C6C-PLH, a new COM Express Compact Type 6 based on Intel Core Ultra Series 3 processors.

Designed for edge AI, robotics, industrial automation, medical imaging, and advanced visualisation systems, the Tria C6C-PLH module gives system designers a standards-based platform for applications that require high computing performance, integrated AI acceleration, and long-term lifecycle stability.

Performance and edge AI

The C6C-PLH is available with Intel Core Ultra Series 3 processor options offering up to 16 cores and Intel Iris Xe graphics with up to 12 Xe cores. By combining CPU, GPU, and integrated neural processing capabilities on a compact 95 mm x 95 mm module, the new platform helps embedded design teams bring AI enabled systems closer to the point of operation, which significantly reduces dependence on cloud connectivity while supporting real time analytics, machine vision, predictive maintenance, and intelligent control.

For customers building industrial equipment, autonomous machines, HMI platforms, or image intensive medical systems, the module is designed for professional applications at the edge, balancing performance with the practical requirements of embedded deployment. Suitable for a range of industrial temperatures, it supports up to 96 GB of LPDDR5x memory, with in-band ECC options, four independent display streams, and a broad set of high-speed interfaces, including PCI Express Gen 4, rich USB options such as USB4, and 2.5 Gb Ethernet based on Intel i226.

"Industrial customers are not simply looking for more processing power; they need a platform that can sustain demanding workloads throughout long product lifecycles," said Daniel Denzler, Vice President of Tria Technologies. "The COM Express form factor gives customers a proven foundation on which to build with confidence, while the C6C-PLH brings the processing, graphics, AI acceleration, and industrial-grade options those systems require."

Key benefits of the Tria C6C-PLH module:

·       Accelerates edge AI with integrated CPU, GPU, and NPU performance.

·       Enables real-time processing with less reliance on the cloud.

·       Simplifies integration with rich connectivity and COM Express Type 6 compatibility.

·       Supports dependable, long-life industrial deployment.

The integrated NPU accelerates visual inspection, anomaly detection and other machine-vision workloads at the edge. Further AI acceleration is built into the Intel Xe graphics processor units. Intel Iris Xe graphics supports multiple displays, advanced operator interfaces and media-rich visualisation, while hardware acceleration for media transcoding and rendering extends the module’s suitability for imaging and video applications.

Built for industrial deployment

Industrial features include support for Intel Time Coordinated Computing and Time Sensitive Networking, industrial temperature variants from -40°C to +85°C, 24/7 operation, and TPM 2.0 security. The module also supports Windows 11 IoT Enterprise LTSC 2024 and a Linux BSP based on Yocto, helping developers align the platform with existing software strategies and long-term maintenance requirements.

Standards-based integration and support

The Tria C6C-PLH uses the COM Express Type 6 pinout, allowing customers to draw on an established ecosystem of carrier boards and accessories while retaining flexibility for custom system designs. Tria also offers evaluation carrier boards and cooling solutions to support development, prototyping, and deployment across different mechanical and thermal environments.

"The challenge for embedded teams is turning a successful AI demonstration into a secure, scalable product that can be maintained for years," added Denzler. "Tria combines a proven modular approach with embedded design expertise and long-term product availability to help customers make that transition."

Further information can be found here.