RK3588S Single Board Computer
ArmSoM-
Sige7s
Flagship AI Single Board Computer — 8nm Octa-Core, 6 TOPS NPU, WiFi 6 + Gigabit Ethernet, Ready to Use
8 Cores
A76 + A55 @ 2.4GHz
6 TOPS
NPU AI Compute
8K@60
Video Codec
32GB
Max Memory
Six Core Features
Flagship SBC based on RK3588S with integrated WiFi 6, Gigabit Ethernet, 40pin expansion, ready to use out of the box
⚡
8-Core 64-bit CPU
4× Cortex-A76 @ 2.4GHz + 4× Cortex-A55 @ 1.8GHz, 8nm advanced process, powerful multi-threaded performance
8nm | 8-Core | 2.4GHz
📡
WiFi 6 + Gigabit Ethernet
Onboard IEEE 802.11a/b/g/n/ac/ax WiFi 6 + BT 5.0 (AIC8800), Gigabit Ethernet port with PoE support
LPDDR4/4X | 32GB | eMMC 256GB
🤖
6 TOPS NPU
Built-in AI accelerator supporting INT4/INT8/INT16 mixed-precision, easily deploy YOLO, ResNet and other mainstream models
INT4/INT8/INT16 | TensorFlow | PyTorch | ONNX
🔌
40pin Compatible + Rich Interfaces
Raspberry Pi 40pin compatible, 2×USB 3.0 + 2×USB 2.0, HDMI 2.1 + DP 1.4 + MIPI CSI/DSI, 5V~23V wide voltage input
RPi 40pin | USB 3.0×2 | HDMI 2.1
🎮
Mali-G610 MP4 GPU
Quad-core GPU supporting OpenGL ES 3.2 / OpenCL 2.2 / Vulkan 1.2, smooth 4K UI and 3D rendering
450 GFLOPS | Vulkan 1.2
🎥
8K Full-Pipeline Video
8K@60fps decode (H.265/VP9/AVS2), 8K@30fps encode (H.264/H.265), HDMI 2.1 direct to 8K displays
Decode 8K@60 | Encode 8K@30
Product Showcase
85×56mm compact board integrating RK3588S SoC + LPDDR4X + eMMC + WiFi 6 + Gigabit Ethernet
Sige7s Board Front
RK3588S SoC | LPDDR4X RAM | eMMC Storage | WiFi 6 Module | Gigabit Ethernet
Interface Layout
HDMI 2.1 | DP 1.4 | USB 3.0×2 | USB 2.0×2 | Gigabit Ethernet | 40pin Header | MIPI CSI/DSI
RK3588S — Flagship Chip, Compact Form
RK3588S is the miniaturized and cost-optimized variant of the RK3588 family. Same core computing performance, smaller package, ideal for compact boards
RK3588S vs RK3588 Quick Comparison
RK3588S retains the core compute units (CPU/GPU/NPU/codec) while reducing peripheral interfaces for a smaller package and lower cost. For applications that don't require multi-lane PCIe 3.0, HDMI RX, or dual GMAC, RK3588S is the better choice.
RK3588S
8nm Flagship AIoT SoC | 17×17mm Package
CPU
A76×4 + A55×4
GPU
Mali-G610 MP4
Process
8nm
Max Frequency
2.4 GHz
NPU
6 TOPS
Package
17×17mm
Specification | RK3588 | RK3588S |
CPU | A76×4 + A55×4 | A76×4 + A55×4 |
Max Frequency | 2.4GHz | 2.4GHz |
GPU | Mali-G610 MP4 | Mali-G610 MP4 |
NPU | 6 TOPS | 6 TOPS |
8K Decode | 8K@60fps | 8K@60fps |
8K Encode | 8K@30fps | 8K@30fps |
Package Size | 23×23mm | 17×17mm ▼ 45% |
PCIe 3.0 | 2×2-lane | None |
PCIe 2.0 | 3×1-lane (PCIe 2.1) | 2×1-lane (PCIe 2.1) |
SATA 3.0 | 3 ports | 2 ports (muxed with PCIe 2.0) |
HDMI TX / eDP | 2× HDMI 2.1 TX / eDP 1.3 | 1× HDMI 2.1 TX / eDP 1.3 |
HDMI RX | 1× 4-lane HDMI 2.0 RX (up to 4K@60fps) | Not supported |
DP Output | 2× 4-lane DP 1.4 (muxed with USB 3.0) | 1× 4-lane DP 1.4 (muxed with USB 3.0) |
MIPI CSI | 2× 4-lane or 4× 2-lane | 1× 4-lane or 2× 2-lane |
Full-featured GMAC | 2 (GMAC0 + GMAC1), expandable dual Gigabit Ethernet | 1 (GMAC1 only), single Gigabit Ethernet |
ADC | 8× SARADC | 6× SARADC |
● Core computing performance is identical
● Peripheral interfaces are simplified
Performance Comparison
Sige7s leads in AI compute, CPU performance, and video codec across similarly priced SBCs
AI Compute Comparison
Product | AI Accelerator | AI Compute | Notes |
ArmSoM Sige7s | Built-in NPU | 6 TOPS | INT4/INT8/INT16 mixed-precision |
Orange Pi 5 | Built-in NPU (RK3588) | 6 TOPS | Equal to Sige7s |
Raspberry Pi 5 | No built-in NPU | — | CPU/GPU inference |
Jetson Nano | No built-in NPU, relies on GPU | ~0.5 – 0.512 TFLOPS | 128-core Maxwell GPU inference |
CPU Multi-Core Performance (Relative)
ArmSoM Sige7s
A76×4 + A55×4
Orange Pi 5
A76×4 + A55×4 @ 2.4GHz
Raspberry Pi 5
A76×4 @ 2.4GHz
Jetson Nano
A57×4 @ 1.43GHz
Video Decode Capability
ArmSoM Sige7s
8K@60fps
Orange Pi 5
8K@60fps
Raspberry Pi 5
4K@60fps
Jetson Nano
4K@30fps
Edge AI Inference
6 TOPS NPU supports mainstream deep learning frameworks — from prototype to deployment in one solution
Built-in NPU supports INT4 / INT8 / INT16 mixed-precision with up to 6 TOPS
👓
Computer Vision
Face detection, object recognition, behavior analysis, industrial inspection
🤖
Robotics
SLAM navigation, path planning, sensor fusion, edge decision-making
🎧
Intelligent Voice
Voice wake, voiceprint recognition, noise reduction, TTS
📷
Smart Security
Intrusion detection, license plate recognition, people counting, video structuring
Multi-Channel I/O, Multi-Screen Display
RK3588S supports simultaneous independent multi-display output for digital signage, HMI, and multi-screen workstation scenarios
8K@60fps
HDMI 2.1
7680×4320
8K@30fps
DP 1.4
DisplayPort Alt Mode
4K@60fps
MIPI DSI (2-Lane)
Embedded Display Interface
5MP
MIPI CSI (1×4-lane or 2×2-lane)
Camera Input
Supports simultaneous independent HDMI + DP output for digital signage, HMI, and multi-screen workstation scenarios
Rich Interfaces, Unlimited Expansion
ArmSoM-Sige7s onboard 40pin Raspberry Pi compatible header + USB 3.0 + Gigabit Ethernet + HDMI 2.1 + MIPI CSI/DSI, ready to use
🖥
HDMI 2.1
8K@60fps video outpu
🖥
DP 1.4
DP 8K 30fps
🔌
USB 3.0
2× USB 3.0 (5Gbps)
🔌
USB 2.0
2× USB 2.0 (480Mbps)
🌐
Gigabit Ethernet
GbE + PoE power support
📡
WiFi 6 + BT 5
AIC8800 wireless module
⚙
SPI / PWM
1×SPI + 2×PWM
💡
Status LEDs
2× status indicators
📷
MIPI CSI
1×2-lane, supports 5MP camera input
🔋
RTC Battery
LK8563S low-power RTC
📺
MIPI DSI
2-Lane display output
⚙
CAN
1×CAN
📏
40pin Header
Raspberry Pi 40pin compatible
⚙
ADC
6× SARADC
⚙
UART / I2C
3×UART + 2×I2C
🔘
Buttons
PWRON + Maskrom
Use Cases
From ARM PC to edge AI, from smart NVR to robotics, Sige7s empowers every industry
📺
ARM Personal Computer
8-core CPU + 32GB RAM, runs Ubuntu desktop, handles everyday office and development
🌐
Edge Server
Gigabit Ethernet + PoE power, lightweight cloud server, blockchain node
🤖
Edge AI & ML
Deploy AI models locally, 6 TOPS NPU low-latency inference, protect data privacy
🛠
Industrial HMI
Touchscreen HMI, 40pin expansion CAN/SPI, industrial field control
💻
Digital Signage
8K video playback + HDMI 2.1 + DP dual-screen output, drives commercial display systems
📷
Machine Vision
MIPI CSI camera + NPU inference, industrial inspection, object recognition
🎥
Smart NVR
Multi-channel video input + AI analysis, smart recording and alerts, WiFi 6 wireless backhaul
🎮
Media Center
8K@60 decode capability, HDMI 2.1 direct to 8K TV, build a home theater
Technical Specs
Complete hardware parameters for carrier board design and product integration
Item | Specification |
|---|---|
SoC | Rockchip RK3588S, 8nm process |
CPU | 4× Cortex-A76 @ 2.4GHz + 4× Cortex-A55 @ 1.8GHz |
GPU | ARM Mali-G610 MP4 (4×256KB L2 Cache), supports OpenGL ES 3.2 / OpenCL 2.2 / Vulkan 1.2 |
NPU | 6 TOPS@INT8 (3 NPU cores), supports INT4/INT8/INT16 mixed-precision |
VPU / Codec | Hardware decode: 8K@60fps H.265/VP9/AVS2, 8K@30fps H.264, 4K@60fps AV1, 1080P@60fps MPEG-2/VC-1/VP8; Hardware encode: 8K@30fps H.265/H.264 |
ISP | Integrated 48MP ISP, supports HDR & 3DNR |
Memory | LPDDR4/4X, options: 8GB / 16GB / 32GB (max 32GB) |
Storage | Onboard eMMC, options: 16GB / 32GB / 64GB / 128GB; MicroSD card expansion supported |
Networking | WiFi 6 + BT 5.0 (AIC8800); Gigabit Ethernet port, supports PoE power (requires additional PoE module) |
Video Output | 1× HDMI 2.1 TX / eDP 1.3 (8K@60fps); 1× DP 1.4 via Type-C (4K@60fps, muxed with USB 3.0); 1× MIPI DSI (2-Lane, 1280×800) |
Video Input | 1× MIPI CSI (4-Lane or 2× 2-Lane, up to 5MP camera) |
Audio | HDMI audio output; HP-OUT headphone output; USB Type-C (DP1.4) audio output |
USB | 2× USB 3.0 (USB 3.1 Gen1, 5Gbps); 2× USB 2.0 (480Mbps) |
PCIe / SATA | 2× PCIe 2.1 (1-lane); 2× SATA 3.0 (muxed with PCIe 2.0) |
Ethernet | 1× GMAC (GMAC1 only), single Gigabit port, supports PoE (requires additional PoE module) |
40pin Header | Fully Raspberry Pi 40pin compatible, supports UART / SPI / I2C / I2S / PWM / 6× SARADC / 5V / 3.3V power |
Other | RTC battery connector (LK8563S); 2× status LEDs; PWRON button + Maskrom button |
Power Supply | Wide voltage input 5V ~ 23V (±5% tolerance); recommended 5V/3A USB Type-C |
PCB Layers | 12 layers |
Dimensions | 85 mm × 56 mm × 17 mm |
Weight | 46.6g |
Operating Temperature | 0°C ~ 80°C |
OS | Official: Debian 12, Ubuntu 22.04, Ubuntu 24.04; Third-party: Armbian |
Compatible Ecosystem
Raspberry Pi 40pin compatible ecosystem with rich accessories, lowering the development barrier
🔌
Raspberry Pi 40pin
Fully compatible with RPi 40pin header and HAT expansion boards
📷
RPi Camera
Compatible with Raspberry Pi IMX219 MIPI CSI camera module
🖥
RPi 7" Display
Compatible with Raspberry Pi 7-inch touchscreen display
🔧
Cooling Solution
Compatible with Raspberry Pi 4 cooling, supports active cooling
Design Resources
Open documentation with complete materials, from product docs to source code, accelerating development
FAQ
Q
What is the difference between Sige7s and Sige7?
Sige7 is based on RK3588 (full version), while Sige7s is based on RK3588S (lite version). Core computing performance (CPU/GPU/NPU/codec) is identical. RK3588S reduces some peripheral interfaces (no PCIe 3.0, fewer HDMI/DP/SATA channels), smaller package (17×17mm vs 23×23mm), lower cost. For most applications, Sige7s offers the same compute power at better value.
Q
What is the relationship between Sige7s and CM7s?
Both Sige7s and CM7s are based on the Rockchip RK3588S chip with identical core computing performance (CPU/GPU/NPU/codec). The main difference is the product form: CM7s is a 55×40mm B2B compute module that requires a carrier board; Sige7s is an 85×56mm complete single-board computer with onboard WiFi 6, Gigabit Ethernet, USB ports, etc., ready to use out of the box. Both share the same software ecosystem and AI capabilities but target different use cases: CM7s for embedded product integration, Sige7s for rapid prototyping and standalone applications.
Q
Is Sige7s compatible with Raspberry Pi accessories?
Yes, the Sige7s onboard 40pin header is fully compatible with the Raspberry Pi 40pin definition, supporting UART / SPI / I2C / I2S / PWM / ADC interfaces. It can connect to Raspberry Pi HAT expansion boards, camera modules (recommend IMX219), and LCD displays (recommend Waveshare 7" 800×480).
Q
What operating systems does Sige7s support?
Official support: Debian 12, Ubuntu 22.04, and Ubuntu 24.04. Community support: Armbian. The ArmSoM team has tested and verified these systems. Pre-built images can be downloaded from Google Drive.
Q
How to flash the OS onto Sige7s?
Two methods: 1) eMMC flashing: Use a USB A-to-A cable to connect the onboard USB 3.0 port to a PC, and use RKDevTool to flash; 2) SD card boot: Use SD_Firmware_Tool to write the image to a MicroSD card (Class 10 or above, at least 8GB).
Q
What is the power supply for Sige7s?
Sige7s supports wide voltage input 5V ~ 23V (±5% tolerance), recommended 5V/3A USB Type-C power supply. The Ethernet port also supports PoE power (requires additional PoE module). The power port is labeled "PWR IN".
Q
How to deploy AI models on the NPU?
Use the Rockchip RKNN Toolkit to convert models from TensorFlow, PyTorch, ONNX, etc. to RKNN format, then run on the NPU. ArmSoM provides RKNN-LLM documentation and example code at docs.armsom.org.
Ready to use Sige7s in your project?
From sample procurement to mass customization, ArmSoM provides full-chain support
Or email sales@armsom.org