Advantech MIO-5355 — 3.5-inch SBC on Qualcomm QCS6490 / QCS5430 Edge AI

Specifications

Overview

The Advantech MIO-5355 is a 3.5-inch industrial SBC offered on either the Qualcomm Dragonwing QCS6490 or the QCS5430. The two-platform approach lets a product family span an octa-core Kryo 670 part with a 12.3 TOPS AI engine and a hexa-core part rated at 3.5 TOPS while keeping the same mechanical and I/O design.

The QCS6490 configuration runs one Kryo 670 Gold Plus core at 2.7 GHz, three Gold cores at 2.4 GHz and four Silver cores up to 1.9 GHz, with an Adreno 643 GPU, an Adreno 633 VPU handling 4K60 decode and 4K30 encode, a Spectra ISP reaching 64 MP or triple 22 MP at 30 fps zero-shutter-lag, and a 6th-generation Qualcomm AI Engine combining the Hexagon DSP with dual HVX, the Hexagon Co-processor 2.0 and a Hexagon Tensor accelerator for 12.3 TOPS. The QCS5430 variant trims to two Gold Plus and four Silver cores at 2.1 GHz and 1.8 GHz, uses an Adreno 642L, and its dual Spectra ISP supports 22 + 22 MP cameras with 720p at 960 fps slow motion.

Both variants take up to 8 GB of soldered LPDDR5 and up to 128 GB of UFS or eMMC flash, with an M.2 B-Key 2280 socket for additional NVMe storage. Advantech supports Windows 11 IoT Enterprise and Ubuntu 24.04 LTS. Notably, other QCS6490 hardware on the market tends to be compact or maker-oriented; the MIO-5355's differentiator is that it puts the same SoC on a standard industrial 3.5-inch board with the mounting pattern, power input and long-lifecycle sourcing that industrial integrators expect.

Key Benefits

Two Qualcomm tiers (12.3 TOPS QCS6490, 3.5 TOPS QCS5430) on one 3.5-inch design lets a product family scale performance without a mechanical redesign. Standard industrial 3.5-inch form factor rather than a maker-style board, so it drops into existing enclosures and DIN mounting. Strong ISP capability for multi-camera vision — up to 64 MP single or triple 22 MP at 30 fps ZSL. Up to 8 GB LPDDR5 and 128 GB UFS plus M.2 2280 NVMe provides usable storage, and both Windows 11 IoT Enterprise and Ubuntu 24.04 LTS are supported.

Applications

Industrial edge AI inspection and vision, retail and hospitality smart terminals, medical and laboratory instruments with imaging, smart-city and access-control endpoints, robotics and AGV controllers, digital signage with 4K decode.

Related Products