High-Performance Application Compute
8-core Kryo Gen 6
- Dual quad-core clusters
- Up to 2.36GHz
- Built for compute-intensive applications
Qualcomm IQ-9075
200 TOPS
36GB LPDDR5
UFS 3.1
AI Agent
Ubuntu 24.04
Robotics R&D
LLM / VLM
Sensor I/O
AI Toolchain
Display
2x 2.5GbE
Powered by the Qualcomm Dragonwing IQ-9075 Processor
Designed for demanding robotics and edge intelligence, the IQ-9075 combines high-performance application compute, on-device AI acceleration, advanced graphics, and real-time control on a unified platform.
On-Device AI Engine
Dual Qualcomm Hexagon Tensor Processors
- Up to 100 Dense TOPS / 200 Sparse TOPS
- HVX vector and HMX matrix acceleration
- Built for vision, multimodal, and generative AI workloads
Real-Time Control Subsystem
Quad-core Arm Cortex-R52
- Up to 1.85GHz
- Configurable in lock-step or split mode
- Supports independent boot via OSPI
Advanced Graphics & GPU Compute
Qualcomm Adreno 663
- Up to 840MHz
- Supports Safe GPGPU Compute
- Supports Adreno NN Direct
Vulkan® 1.2
OpenGL ES 3.2
OpenGL 2.0 FP8K Video & Multi-Camera Vision
The Qualcomm Spectra 690 ISP and Adreno VPU 765 enable 8K video decoding, multi-stream 4K processing, and dual-camera vision development.
Up to
8K@60fpsVideo Decode
8K@60fpsVideo Decode
Up to
4× 4K@60fpsMulti-Stream Decode
4× 4K@60fpsMulti-Stream Decode
Up to
2× 4K@60fpsVideo Encode
2× 4K@60fpsVideo Encode
4-lane + 2-laneDual MIPI CSI
Video Codec SupportDecode: AV1 · H.264 · H.265 · VP9 · MPEG-2Encode: H.264 · H.265 · HEIF · HEIC
High-Capacity Memory & Fast Storage
High-bandwidth ECC memory, fast onboard storage, and PCIe Gen4 NVMe expansion support demanding AI and vision workloads.
36GB LPDDR5 Memory
- Up to 6400 MT/s
- Six-channel, 96-bit architecture
- Inline ECC for enhanced data integrity
Onboard 128GB UFS
- UFS 3.1 Gear 4 ×2
- Fast system boot and application loading
M.2 M Key 2280
- PCIe Gen4 NVMe expansion
- High-speed, scalable storage capacity
Qualcomm AI Hub
Documentation, optimized models, and deployment tools to accelerate your on-device AI journey.
AI Agents
AI Models
AI & Robotics Stack
Ready for Modern AI and Robotics Development
Rich Connectivity for Intelligent Devices
40-Pin GPIO Header4x CAN\n(3-Pin 1.25mm Pitch)2x RS232\n(3-Pin 1.25mm Pitch)1x RS485\n(3-Pin 1.25mm Pitch)XT60 Power Connector\n(19–24V Input)314-Pin MXM 3.0 ConnectorDP4x USB 3.2 Type-A\n(10 Gbps)2x RJ45 Ethernet Ports\n(2.5GbE)HDMIDC5525 Power Jack\n(19-24V Input)FAN Connector\n(2-Pin 1.25mm Pitch)USB-C for DebugPower ButtonUSB 3.0\n(8-Pin 1.25mm Pitch)3x USB 2.0\n(4-Pin 1.25mm Pitch)RTC Battery Connector4-lane MIPI DSI\n(39-Pin 0.3mm Pitch)Nano SIM SlotM.2 B Key 3042 SloteDP4-lane MIPI CSI\n(31-Pin 0.3mm Pitch)2-lane MIPI CSI\n(15-Pin 1.0mm Pitch)M.2 M Key 2280 SlotEDL ButtonM.2 E Key 2230 SlotVMARC-Q9075
Applications & Use Cases
VMARC-Q9075 paired with the Radxa C201 VMARC IO Board enables rapid validation and prototype development
Service Robots
Autonomous Mobile Robots
Smart Factory
Drones
Local AI Assistants
Machine Vision
Edge AI & Generative AI
Autonomous Driving
Education & Research
Developer Kit Components
Package List
- 1x Radxa VMARC-Q9075
- 1x Radxa C201 VMARC IO Board
- 1x Wireless Module (Wi-Fi 6 & BT 5.3)
- 1x Active Heatsink
- 1x Protective Base
- 1x Power Adapter
Comprehensive Technical Documentation and
Active Community
Comprehensive Technical Documentation and Active Community !
Radxa provides comprehensive technical documentation, empowering users to fully utilize their products. At the same time, they actively maintain an engaged community, fostering communication and collaboration among users.
Attention:
*Qualcomm branded products are products of Qualcomm Technologies, Inc. and/or its subsidiaries.
The terms HDMI, HDMI High-Definition Multimedia Interface, HDMI Trade dress and the HDMI Logos are trademarks or registered trademarks of HDMI Licensing Administrator, Inc.
