APLUX

Rhino Core IQ9

A flagship industrial System-on-Module powered by Qualcomm QCS9075M (IQ-9075), delivering 100 TOPS INT8 dense compute and dual-engine hard real-time control for high-end robotics, automotive domain control and industrial vision.

QCS9075M100 TOPS INT865×65mm BGA-40~85℃15~25WUbuntu24.04

Highlights

Flagship Compute, Hard Real-Time

100 TOPS INT8 dense compute, 8-core Kryo Gen6 and 4-core Cortex-R52, with multi-stream 4K video processing.

Full Industrial I/O

4× MIPI-CSI, PCIe Gen4, dual USB 3.1, 8× CAN-FD, dual Gigabit TSN and 200+ GPIO.

Industrial-Grade Reliability

-40~85℃, full JEDEC reliability testing, industrial PMIC and a 10-year lifecycle.

Effortless AI Deployment

AIMO graphical model conversion, a complete Ubuntu BSP and 450+ ready-to-use industrial models.

Specifications

Chipset model
Qualcomm QCS9075M (Dragonwing IQ-9075)
CPU
8-core Kryo Gen6 (dual quad-core clusters), up to 2.361GHz
Real-time kernel
4-core Cortex-R52, up to 1.85GHz
AI compute
Hexagon tensor processor (HVX + HMX), up to 100 TOPS (INT8 dense) / 200 TOPS (INT8 sparse)
GPU
Adreno 663, up to 840MHz, supports Vulkan 1.2, OpenGL ES 3.2, OpenCL 2.0 FP
Memory (LPDDR5)
36GB (standard)
Storage
UFS 3.1 / eMMC storage, capacity configured per project
Operating system
Linux、Ubuntu
Video processing (VPU)
Decode: 4× 4K@60fps / 4K@240fps, supports AV1/HEVC/H.264 Encode: 2× 4K@60fps / 4K@120fps, supports HEVC/H.264
Camera interface
4× MIPI CSI 4-lane camera interfaces (D-PHY / C-PHY)
PCIe
1×PCIe 4-lane Gen4 1×PCIe 2-lane Gen4
USB
2×USB 3.1 Gen2 1×USB 2.0
Wireless & network
Wired: 2× SGMII Gigabit Ethernet (AVB/TSN) Wireless: Wi-Fi 6E / Bluetooth 5.3 (optional)
Industrial control
CAN bus: 8× CAN-FD (up to 10Mbps) Serial: multiple UART ports, RS485 supported GPIO: 149 in the main domain + 78 in the real-time domain (1.8V) Other: I2C, SPI, I2S/PCM/TDM
Display interface
2× MIPI DSI + 4× eDP/DP 1.4, up to 5× 4K outputs
Environment
Operating temperature: -40℃~+85℃ Storage temperature: -40℃~+125℃
Power input
DC 3.3V input (wide-range carrier board support)
System power
Typical full-load power: 15W~25W
Package size
65×65mm BGA 1573-ball surface-mount package
Product lifecycle
10+ years

Build with Rhino Core IQ9

Get Rhino Core IQ9 documentation, evaluation kits, pricing and support for your Physical AI project.