To accelerate Qualcomm Technologies’ leadership in AI and the Connected Intelligent Edge, we are unifying our existing best-in-class AI software offerings into a single package, the Qualcomm AI Stack. This will help our OEM customers and developers create, optimize, and deploy their AI applications on Qualcomm Technologies products and fully leverage our Qualcomm AI Engine performance. 

The Qualcomm AI Stack, from top to bottom, supports popular AI frameworks such as TensorFlow, PyTorch, ONNX and runtimes including TensorFlow Lite, TensorFlow Lite Micro, ONNX runtime.  In addition, it includes inferencing SDKs like our popular Qualcomm Neural Processing SDK with the recently announced Windows version. Our developer libraries and services support the latest programming languages, virtual platforms based on open-source QEMU, and compilers including LLVM and TVM. At a lower level, our system software includes the basic RTOS, system interfaces and drivers. We also have a rich variety of OS support, spanning across different product lines, support includes Android, Windows, Linux and QNX and deployment and monitoring infrastructure like Prometheus, Kubernetes and Docker. These combinations of OS support will enable advanced features, for example, digital twins for automotive use cases that enable real-time data on both the device and in the cloud. 

The Qualcomm Neural Processing SDK for Android and the newly announced Windows version remains a key product for OEMs and developers to run their neural networks on our AI accelerators that supports a broad range of Qualcomm Technologies products and will continue to receive regular updates and support.       

As part of the new offering, the Qualcomm AI Engine direct will now scale across every AI accelerator inside a broad range of products. OEMs and developers now will have the ability to develop and optimize AI models using the Qualcomm AI Stack once, then move the same model across different Qualcomm Technologies products and tiers for a true “develop once, deploy anywhere” model that helps OEMs and developers save time and resources.

Qualcomm AI Stack can be used directly by OEM and developers to develop on mobile and compute products.  It is also a building block for our domain specific SDKs that are available today, for example, the Snapdragon Ride SDK, the Intelligent Multimedia SDK and the Snapdragon Spaces SDK.  All these SDKs are powered by the Qualcomm AI Stack so any AI models that are already developed on the Qualcomm AI Stack can be easily loaded and deployed onto these domain specific SDKs for faster development.

The Qualcomm AI Stack will also include our popular tools, including Qualcomm AI Model Efficiency Toolkit (AIMET) Pro, AI Model Efficiency Toolkit (AIMET), Model Zoo, Model Analyzers and Neural Architecture Search (NAS) which includes Google Cloud Vertex AI NAS that we announced at Snapdragon Tech Summit 2021.

The Qualcomm AI Stack revolutionizes how we think about AI software and provides the ultimate tools for our OEM and developers to create faster and smarter AI applications.

Qualcomm AI Stack is available today with a full roadmap of features coming in the future. 

Qualcomm AI Stack, Qualcomm AI Engine, Qualcomm Neural Pro SDK, Snapdragon, Snapdragon Ride, Snapdragon Spaces, and Qualcomm AIMET Pro are products of Qualcomm Technologies, Inc. and/or its subsidiaries. AIMET is a product of Qualcomm Innovation Center, Inc.