The Radxa C200 looks like a single-board computer, although it is more of a developer kit with a different focus point. The device is based on the Nvidia Jetson Orin NX 8GB and sports six ARM Cortex-A78AE cores, 1,024 CUDA as well as 32 Tensor cores. This makes the SBC particularly well suited for running AI models locally.
One possible scenario would be local image recognition in industrial environments. The maximum performance is listed at 117 TOPS. This comparatively high performance becomes evident when looking at the active cooling system, while other single-board computers like the Raspberry Pi 5 are often passively cooled, depending on the use case. Either way, the Radxa C200 is also equipped with 8GB LPDDR5 RAM.
Furthermore, there are several M.2 slots available, specifically one in the M.2 2280 and two in the M.2 2230 format, so users can connect multiple SSDs at once. Depending on the M.2 slot, up to four PCIe 4.0 lanes are supported. In addition, there are some typical connectivity options for SBCs, such as four USB 3.2 Gen 2 Type-A ports and a single USB 3.2 Gen 2 Type-C port.
Network access can be provided via Gigabit Ethernet, but a WiFi card can be installed as an alternative. DisplayPort 1.2 is also supported, as are some features that are not typically found in consumer devices. These include MIPI CSI, a 40 pin GPIO and headers for PoE and CAN. The new Nvidia-powered SBC from Radxa is now available at Arace for $499.




















