A system-on-module that can survive motion, vibration, and extreme temperatures is now available for developers building AI-driven edge devices. SolidRun’s latest COM Express Type 6 modules—codenamed P100—combine AMD’s Ryzen AI embedded processors with a unique memory architecture designed to balance serviceability and durability in mobile platforms.
The P100 series addresses a growing need for edge AI hardware that can operate reliably in non-static environments, such as rail systems, autonomous mobile robots (AMRs), or industrial automation equipment. Traditional soldered memory modules often fail under continuous movement, but SolidRun’s LP-CAMM2 integration offers the mechanical robustness of solder while retaining the upgradeability of modular memory. This approach also supports high-bandwidth LPDDR5X-class speeds (up to 9600 MT/s), critical for real-time sensor processing and AI inference.
- Processor: AMD Ryzen AI Embedded P100
- Memory: LP-CAMM2 with LPDDR5X (up to 9600 MT/s)
- AI Performance: Up to 50 TOPS from NPU, 59 TOPS total SoC
- Power Envelope: Configurable 15 W–54 W
- Thermal Range: -40°C to +85°C (no fan required)
The modules are built around AMD’s latest CPU, GPU, and NPU architectures, delivering embedded-friendly power profiles while pushing performance for real-time vision tasks. The 15 W–54 W power range allows for fanless designs in sealed or high-performance systems, making them suitable for applications like medical carts, heavy machinery, or aerospace defense where longevity matters as much as compute.
For developers, SolidRun offers a HoneyComb evaluation platform to prototype and test the P100 modules before integrating into custom carriers. The focus on serviceability—through screw-lock retention of memory modules—means field repairs become more practical without sacrificing mechanical resilience in moving systems. This balance positions the P100 series as a key enabler for AI at the edge, particularly where motion and durability are non-negotiable.
