A green circuit board labeled AMD-FP8-COM6 with a visible CPU socket and a Crucial 64GB memory module.

SolidRun Unveils Edge AI Modules Powered by AMD Ryzen AI Embedded P100 SoCs with LPCAMM2 Memory

SolidRun has unveiled a new lineup of edge AI computing modules built for demanding, real-world deployments: the P100 COM Express Type 6 family, powered by AMD Ryzen AI Embedded P100 processors and paired with LPCAMM2 memory. Designed for systems that operate in motion and in harsh conditions, these modules target industrial automation, transportation, medical devices, robotics, and other mission-critical edge environments where reliability and serviceability matter as much as performance.

One of the standout features is SolidRun’s integration of LP-CAMM2 memory into an industrial COM Express Type 6 module. This approach combines the practical benefits of replaceable, upgradeable memory with the durability typically associated with soldered solutions, while keeping power consumption low thanks to LPDDR5-class efficiency. For OEMs and engineers, that means memory can be serviced or swapped in the field without sacrificing the mechanical stability needed for mobile platforms.

LP-CAMM2 also brings advantages that are especially relevant to devices exposed to constant vibration and movement. The more secure mechanical retention, including screw-lock fastening, helps reduce connector movement over time—an issue that can shorten lifespan or create intermittent failures in equipment mounted on vehicles or operating on factory floors. At the same time, the platform is built for high bandwidth, offering LPDDR5X-class speeds in the 8533 to 9600 MT/s range, supporting fast AI inference and real-time sensor processing.

On the AI side, the P100 COMx6 series is aimed at workloads such as machine vision, real-time analytics, and autonomous or semi-autonomous control. With AMD’s latest CPU, GPU, and NPU architectures, the modules deliver up to 50 TOPS from the integrated NPU and up to 59 TOPS total AI performance across the SoC. This makes the platform a strong option for edge AI deployments that need near real-time decision-making without relying on cloud connectivity, while staying within embedded-friendly power limits.

SolidRun also emphasizes deployment flexibility and industrial endurance. The modules support an extended operating temperature range from -40°C to +85°C, making them suitable for sealed enclosures, fanless systems, and installations that face high heat, cold climates, or fluctuating environmental conditions. Power can also be configured across a broad envelope from 15W to 54W, helping system designers balance performance, thermals, and energy consumption depending on whether the end product is a compact mobile unit or a higher-performance industrial controller.

To speed up evaluation and prototyping, SolidRun is offering a HoneyComb evaluation platform for the P100 COM Express Type 6 family. The goal is to reduce time-to-market by giving developers a ready-to-use baseline for testing, validation, and real-world application trials, with the option to customize carrier designs to match specific deployment needs.

SolidRun positions the new platform for several high-demand edge AI markets. These include mobile robotics such as AMRs and AGVs that require resilient perception and inference while operating in motion; transportation and rail systems that need rugged computing for monitoring and real-time analytics; industrial automation and machine vision where memory bandwidth and AI acceleration support inspection and closed-loop control; mobile and harsh-environment medical systems like diagnostic carts and connected clinical devices; and heavy machinery or off-highway equipment where durability is critical for sensor-driven automation and predictive maintenance.

SolidRun says the P100 COM Express Type 6 module family and the HoneyComb evaluation platform are available now, along with a software support package and development tools intended to streamline product development and accelerate deployment.