Khadas Edge 2
Ultra-thin 82×57.5×5.7mm RK3588S SBC with WiFi 6, Bluetooth 5.0, USB-C DisplayPort, four independent display outputs, and a 6 TOPS NPU — the thinnest RK3588S single-board computer available.
$229.90
Tap any value to browse matching devices
Specifications
Components
Rockchip RK3588S2 (enhanced RK3588S): 4× Cortex-A76 @ 2.25GHz + 4× Cortex-A55 @ 1.8GHz, Mali-G610 MP4 GPU, 6 TOPS NPU, 8K H.265 decode.
AMPAK AP6275P PCIe WiFi 6 (802.11ax) + Bluetooth 5.0 module with 2×2 MIMO.
GPIO Map
Overview
The Khadas Edge 2 is one of the thinnest single-board computers ever produced — just 5.7mm bare or 11mm with its bundled heatsink and fan. Built around the Rockchip RK3588S2 (an enhanced RK3588S revision), it delivers the full octa-core 2.25GHz ARM performance and 6 TOPS NPU capability of the RK3588 family in a form factor closer to a credit card than a traditional SBC.
The design sacrifices built-in Ethernet and a traditional 40-pin GPIO header to achieve its extreme thinness. Networking requires either a USB dock or the Edge2 IO expansion board. Instead, Khadas focuses on wireless: an integrated AP6275P module provides WiFi 6 (802.11ax, 2×2 MIMO) and Bluetooth 5.0. All expansion is handled through two 30-pin FPC connectors exposing CPU signals (I2C, UART, SPI, I2S, SDMMC, ADC, PWM, USB) and MCU signals, plus 7 pogo pads.
Display support is a highlight: the USB-C port supports DisplayPort Alt Mode alongside USB 3.1 data and USB Power Delivery, enabling four independent simultaneous display outputs. Two MIPI DSI connectors and three MIPI CSI interfaces are exposed via FPC for embedded display and camera integration.
Available in Basic (8GB/32GB) and Pro (16GB/64GB) configurations, the Edge 2 ships as either a standalone SBC (Maker Kit) or as an ARM mini PC with aluminum enclosure. Armbian, Ubuntu, Debian, and Android are supported. It excels as a compact AI inference node, embedded vision platform, and tightly integrated embedded computing module.
Use Cases
6Edge AI
On-device machine learning inference at the edge, without cloud dependency
Computer Vision
Image recognition, object detection, and ML inference using embedded cameras and AI-capable processors.
ARM SBC Development
Used as an ARM Linux development and prototyping platform — running full Debian/Ubuntu for building and testing ARM-native software and embedded Linux applications.
Embedded Linux
Running a full Linux distribution in embedded or headless applications, such as servers, gateways, edge devices, and automation controllers.
Media Player
Building custom media playback devices for audio, video, or digital signage using DVI/HDMI output.
IoT Gateway
Acting as a local IoT gateway that aggregates data from sensors, runs MQTT broker, Node-RED, or Home Assistant, and bridges to the cloud.
Resources
Where to Buy
Explore more of the directory — popular use cases, firmware, and specs to browse next.
