Biostar unveiled its EdgeComp MS-NAT5000, a compact edge AI computing system built around the NVIDIA Jetson Thor T5000 module. The platform targets system integrators and AI developers working on industrial automation and autonomous machines, bringing generative AI capabilities directly to edge environments.

Powered by NVIDIA's Blackwell architecture, the system delivers up to 2,070 TFLOPS of FP4 AI performance with 128GB of LPDDR5X memory and a 14-core Arm CPU. This combination supports LLM, VLM, robotics, and multimodal AI applications while processing data locally to reduce cloud dependency and improve response times.

Connectivity includes a QSFP28 100G interface plus dual 5GbE LAN ports, paired with NVIDIA Holoscan Sensor Bridge for high-speed sensor streaming. Industrial integration comes via RS232/422/485 ports and four CAN FD interfaces, while M.2 slots accommodate NVMe storage, 4G/5G, and wireless modules.

The EdgeComp MS-NAT5000 handles real-time visual inspection, predictive maintenance, and autonomous navigation across manufacturing, logistics, and construction applications. By enabling on-device generative AI and sensor fusion, it gives developers a practical foundation for intelligent edge solutions that operate beyond simple analysis.