The Qualcomm® QCS610 SoC is a high performance IoT System-on-Chip (SoCs) delivering premium features for building advanced smart camera and IoT use cases encompassing machine learning, edge computing, sensor processing, voice UI enablement and integrated wireless connectivity cost effectively.
The QCS610 SoC is engineered to deliver powerful computing for on-device camera processing and machine learning, with exceptional power and thermal efficiency, across a wide range of IoT applications. It integrates a powerful image signal processor (ISP) and the Qualcomm® Artificial Intelligence (AI) Engine, along with a heterogeneous compute architecture including highly optimized custom CPU, GPU and DSP for accelerated AI performance.
Dual 14-bit Qualcomm Spectra™ 250L ISP capable of supporting up to 21MP at 30fps
Qualcomm® Adreno™ 612 GPU with 64-bit addressing @ up to 845MHz with latest API support
Qualcomm® Hexagon™ DSP with dual hexagon vector extensions for running DNN models and advanced Qualcomm® Neural
Processing Engine SDK support
Octa-core Qualcomm® Kryo™ 460 CPU optimized for power and DMIPS
Qualcomm AI Engine designed to support accelerated on-device machine learning
Low power sensor core helps support always-on use cases at reduced power levels
Supports Ethernet, 1x1 802.11ac Wi-Fi and Bluetooth 5.0
Fabricated using the advanced 11nm FinFET process for exceptional thermal and power efficiency
Enhanced Audio Noise and Echo Cancellation Support and Voice UI
HW based security designed with features such as secure boot from hardware root of trust, trusted execution environment, hardware crypto engines, storage security, debug security with lifecycle control, key provisioning and wireless protocol security
Optimized AI Software Deployment using the Qualcomm® Neural Processing SDK for a virtually seamless path for NN deployment.
The modular tool supports varied frameworks like Caffe/Caffe2, TensorFlow/TensorFlow Lite and ONNX for optimized execution utilizing the heterogeneous architecture to achieve desired performance.
Modular and Open Software support for GStreamer, Pulse-Audio and Wayland Weston, TensorFlow Lite allowing customers to easily develop and port their multimedia and machine learning middleware.