Avnet Silica & Tria QCS6490 Vision-AI Development Kit
The QCS6490 Vision-AI Development Kit features an energy-efficient, multi-camera, SMARC 2.1.1 compute module, based on the Qualcomm QCS6490 SoC device.
High-performance cores on the QCS6490 SoC include an 8-core Kryo™ 670 CPU, Adreno 643 GPU, Hexagon DSP and 6th gen AI Engine (13 TOPS), Spectra 570L ISP (64MP/30fps capability) and Adreno 633 VPU (4K30/4K60 enc/dec rates), ensuring that exceptional concurrent video I/O processing performance is delivered.
Edge AI Performance
High-performance cores on the QCS6490
SoC include an 8-core Kryo™ 670 CPU, Adreno 643
GPU, Hexagon DSP and 6th gen AI Engine (13 TOPS).
Multi-Camera Support
Camera Support includes 4x MIPI CSI camera
inputs (2x 22-pin connectors, 2x via SMARC edge)
and up to 64MP / 30fps capability.
Compact & Modular
The kit follows the SMARC 2.1.1 standard,
ideal for scalable embedded designs in
space-constrained environments.
Flexible Connectivity
With USB, CAN-FD, Ethernet, and Wi-Fi 6E,
the kit offers flexible connectivity for seamless
integration across industrial and edge environments.
Cross-Platform OS Support
Additional OS options are available, such as
Windows 11 IoT Enterprise. Qualcomm also supports
Android and Ubuntu Linux on the QCS6490.
Off the shelf, ready to buy...
Want to get your hands on the QCS6490 Vision-AI Development Kit? Fill out our short form, and one of our EMEA experts will be in touch ASAP.
Kit Includes:
- Vision AI Carrier Board with SMARC Qualcomm QCS6490 compute module
- 1x MIPI CSI IMX577 12MP camera
- USB-C PD power supply (9V)
- Extruded aluminium heatsink
Price: €637.16
- Compute Module
- Camera & Display Interfaces
- Connectivity & I/O
- Wireless & Expansion
- Audio Subsystem
- Software & OS Support
SM2S-QCS6490 (Qualcomm QCS6490 SoC)
- 8-core Kryo™ 670 CPU:
- 4x Cortex-A78 @ up to 2.7 GHz
- 4x Cortex-A55 @ 1.9 GHz
- GPU: Adreno 643 @ 812 MHz
- DSP/NPU: 6th Gen Qualcomm AI Engine (13 TOPS)
- ISP: Spectra 570L (supports up to 5 concurrent cameras)
- VPU: Adreno 633 (4K30/4K60 encode/decode)
- Memory: 8GB LPDDR5 SDRAM, 64GB UFS Flash
Camera Support
- 4x MIPI CSI camera inputs (2x 22-pin connectors, 2x via SMARC edge)
- Up to 64MP / 30fps capability
Display Outputs
- 1x Mini DisplayPort 1.4 (up to 4K30)
- 1x MIPI-DSI (4-lane)
- 1x eDP (4-lane)
- 2x USB 3.1, 3x USB 2.0, 1x USB-C OTG
- 2x PCIe Gen3 (1-lane), 1x PCIe Gen3 (2-lane)
- 2x Gigabit Ethernet (RJ45)
- CAN-FD interface
- 40-pin Pi-HAT header
- UART, SPI, I2C, I2S, GPIO
- M.2 Key-E slot: Wi-Fi 6E / Bluetooth 5.4 module (optional)
- M.2 Key-M slot: NVMe SSD storage (optional)
- TPM module (optional)
- 2x PDM microphones
- DA7212 stereo audio codec
- Analog stereo headphone jack + mic/line-in
- Digital audio interface
- Yocto-based Linux BSP with Qualcomm Linux stack (kernel 6.6.x)
- Example AI and multi-camera open-source applications
- Windows 11 IoT Enterprise (available 2H24)
- Android and Ubuntu Linux (dependent on demand)
- Industrial Robots
- Smart Kiosks
- Multi-Camera Security Systems
- Drones/UAVs
- Smart City Infrastructure
- Handheld Scanners
Example Applications in Robotics
- Autonomous Mobile Robots (AMRs): Real-time navigation, obstacle avoidance, and SLAM using multi-camera input and onboard AI.
- Collaborative Robots (Cobots): Vision-based safety monitoring, gesture recognition, and adaptive task execution.
- Inspection Robots: High-resolution image capture and AI-based defect detection in manufacturing lines.
- Service Robots: Interactive kiosks or vending machines with vision and voice interfaces.
- Security Robots: Multi-camera surveillance with facial recognition and anomaly detection.
Example Applications in Smart Kiosks
- Retail & Vending Machines: AI-powered product recommendations, inventory monitoring, and customer engagement.
- Information Kiosks: Real-time facial recognition for personalised content delivery or access control.
- Ticketing & Check-in Stations: Vision-based ID verification and voice-guided interaction.
- Interactive Advertising Displays: Audience analytics and dynamic content adaptation based on viewer demographics.
- Smart City Terminals: Public service kiosks with multilingual voice interfaces and real-time video analytics.
Example Applications in Security Systems
- Smart Surveillance Cameras: AI-powered motion detection, facial recognition, and crowd analytics.
- Perimeter Monitoring: Multi-angle coverage with real-time alerts for unauthorized access.
- Retail & Public Spaces: Customer behavior tracking, theft detection, and queue management.
- Access Control Systems: Vision-based identity verification and entry logging.
- Traffic & Parking Management: License plate recognition and vehicle tracking.
Example Applications in Drones/UAVs
- Autonomous Surveillance Drones: Real-time facial recognition, license plate reading, and anomaly detection.
- Agricultural UAVs: Crop health monitoring using multi-spectral imaging and AI-based analysis.
- Delivery Drones: Vision-based landing zone detection and package verification.
- Search & Rescue UAVs: AI-powered victim detection in disaster zones using thermal and visual feeds.
- Inspection Drones: Infrastructure inspection (e.g., bridges, wind turbines) with defect recognition and reporting.
Example Applications in Smart Cities
- Traffic Management Systems: AI-powered vehicle and pedestrian tracking, congestion analysis, and incident detection.
- Public Safety Surveillance: Multi-camera setups for facial recognition, crowd behaviour monitoring, and emergency response.
- Environmental Monitoring Stations: Vision-based detection of pollution, waste, or hazardous conditions.
- Smart Lighting & Utilities: AI-driven control based on real-time environmental and human activity data.
- Interactive City Terminals: Public kiosks with voice and vision interfaces for navigation, alerts, and services.
Example Applications in Handheld Scanners
- Retail & Warehouse Scanners: Fast barcode scanning, inventory tracking, and product recognition.
- Field Service Devices: OCR for document capture, AI-based diagnostics, and voice-guided workflows.
- Security & Access Control: Facial recognition and ID verification in mobile units.
- Healthcare Scanners: Patient ID scanning, medication verification, and real-time data logging.
- Industrial Inspection Tools: Vision-based defect detection and data capture in manufacturing environments.
Artificial Intelligence
Computer Vision & Recognition Overview
Enable machines to interpret and understand visual information from the real world, mimicking human vision capabilities. Learn more about Avnet Silica's core Computer Vision and Recognition solutions. See applications, articles and more.
Generative AI
Generative AIs are artificial intelligences that are designed to analyse text, images, video and other content formatted for human consumption, and use this data to create additional content that people can read, view, or listen to.
Generative AI at the Edge
Generative AI at the edge is transforming the world around us by enabling low-latency, more secure and real-time AI on devices. Having AI 'on the edge' means personalised applications can run more efficiently without reliance on data centres.
Running LLMs Locally
The large language model (LLM) is the foundation for the revolution in generative artificial intelligence (AI) that delivers natural interaction and even reasoning capability.
Machine Learning
Traditional software comprises instructions that electronic systems follow to perform specified actions. Machine learning inverts that by employing algorithms that instead enable systems to learn as they go along.