Aetina’s CoreEdge MXM Series Takes NVIDIA Blackwell GPUs to Rugged Edge AI
January 14, 2026
News
Aetina Corporation launched its forward looking MXM AI Accelerator Module series accelerated by NVIDIA RTX PRO Blackwell Embedded GPUs. It is part of its CoreEdge MXM AI Module family. By leveraging the NVIDIA Blackwell GPU platform, AI performance has three times more performance power when compared to previous models. The modules integrate innovative AI, ray tracing, and neural graphics technologies.
Aetina’s MX5000B-XA model utilizes the NVIDIA RTX PRO 5000 Blackwell GPU to carry 10,496 CUDA cores, 80 RT cores, and 320 Tensor cores, offering a 40.62 TFLOPS FP32 performance. According to the press release, validation testing demonstrates two times faster LLM inference on the Gemma-2-9b model and 80 percent faster AI inference performance in Geekbench AI benchmarks.
The NVIDIA Blackwell MXM series delivers AI computing throughput enhanced up to three times by supporting FP4 precision, DLSS 4 multi-frame generation technology, and GDDR7 memory with bandwidth up to 896 GB/s, allowing the deployment of advanced AI models such as large language models (LLMs) and vision transformers in areas limited by form factor and power limits.
It has a temperature range of -40°C to +85°C and offers dust- and moisture-resistant conformal coating protection. The solution supports flexible TGP power settings and high-vibration resistant mounting holes on MXM Type B modules for secure installation in high-vibration environments. Aetina utilizes wide-temperature components with high-reliability packaging, extensive thermal cycling testing, cold boot testing, and 100 percent chamber testing before shipment.
In addition, Aetina delivers evaluation kits offering PCIe Gen4 x16 interfaces and HDMI outputs to quicken development time and streamline integration.
Jackal Chen, Director of AI Accelerator Card and GPU Products Division at Aetina commented, "By incorporating NVIDIA Blackwell technology alongside Aetina's extensive engineering expertise, we achieve high-performance AI computing for even the most demanding environments, Additionally, we are prepared for the age of AI agents, multimodal applications, and real-time AI-accelerated computing."
The modules are ideal for autonomous mobile robots (AMRs), unmanned vehicles (UAVs/UGVs), smart manufacturing equipment, and medical and surgical assistance applications.
For more information, visit www.aetina.com.
Chad Cox. Production Editor, Embedded Computing Design, has responsibilities that include handling the news cycle, newsletters, social media, and advertising. Chad graduated from the University of Cincinnati with a B.A. in Cultural and Analytical Literature.
