Aetina Corporation has launched its latest MXM AI Accelerator Module series, powered by NVIDIA RTX PRO Blackwell Embedded GPUs, part of its CoreEdge MXM AI Module family. This new series, utilising the NVIDIA Blackwell GPU platform, achieves up to three times the AI performance of its predecessor, setting a new benchmark for edge AI computing.
These modules adhere to the MXM 3.1 industrial standard, offering compact, efficient, and rugged designs for high-precision, high-density workloads in critical applications such as autonomous mobile robots, unmanned vehicles, smart manufacturing equipment, and medical systems. The series brings the power of generative and physical AI closer to the edge, enabling a comprehensive upgrade of smart applications.
Aetina’s flagship MX5000B-XA model, featuring the NVIDIA RTX PRO 5000 Blackwell GPU, is equipped with 10,496 CUDA cores, 80 RT cores, and 320 Tensor cores, delivering 40.62 TFLOPS FP32 performance. Validation tests show double the speed in LLM inference on the Gemma-2-9b model and an 80% increase in AI inference performance in Geekbench AI benchmarks.
Supporting FP4 precision, DLSS 4 multi-frame generation technology, and GDDR7 memory with up to 896 GB/s bandwidth, the NVIDIA Blackwell MXM series boosts AI computing throughput by up to three times. These advancements facilitate the deployment of complex AI models such as large language models and vision transformers in space- and power-constrained edge environments, offering lower latency and higher throughput for demanding applications.
Aetina’s modules occupy only one-fifth of the board space required by traditional graphics cards while delivering workstation-class performance. They support a wide temperature range from -40°C to +85°C and feature dust- and moisture-resistant conformal coating protection. Additionally, they offer adjustable TGP power settings and high-vibration resistant mounting holes, ensuring secure installation in challenging environments. Aetina also guarantees a 5-year product lifecycle support.
The reliability of these modules is underpinned by rigorous engineering methods, including the use of wide-temperature components and extensive thermal cycling and cold boot testing. This ensures system integrators can deploy with confidence in critical applications. Aetina provides evaluation kits with PCIe Gen4 x16 interfaces and HDMI outputs to accelerate development and simplify integration.
“Aetina’s CoreEdge embedded MXM AI accelerator module series demonstrates our commitment to delivering exceptional performance and reliability,” said Jackal Chen, Director of AI Accelerator Card & GPU Products Division at Aetina. “By incorporating NVIDIA Blackwell technology alongside our engineering expertise, we achieve high-performance AI computing for demanding environments. We offer comprehensive engineering services to streamline system integration, helping customers reduce time-to-market and fully harness edge generative and physical AI.”
The MX5000B-XA and MX500B-QA models are available now, with the MX4000B and MX2000B modules expected to enter mass production in Q1 2026. For more information, visit www.aetina.com.
