Qualcomm’s new AI Stack enables OEMs and developers to utilize a unified, comprehensive AI software capabilities. By doing so, OEMs and developers can now develop and leverage the full AI potential of Qualcomm’s Connected Intelligent Edge products, including, smartphones, IoT, automotive, XR, Cloud and Mobile PC.
At Qualcomm, AI is being strategically viewed as a unique horizontal technology layer cutting across different business lines – from mobile to automotive, and others. With its one technology roadmap, Qualcomm is now bringing unified AI capabilities across all its business lines.
With Qualcomm’s AI strengths and scale, Qualcomm’s OEM base and developers now have the ability to leverage Qualcomm’s AI capabilities and create their own unique features and leverage the same AI model across different products and tiers.
The new Qualcomm AI Stack combines and enhances Qualcomm’s AI software offerings as a comprehensive AI solution for OEMs and developers. The Qualcomm AI stack provides direct access to the Qualcomm AI Engine and dedicated AI cores on Qualcomm Cloud AI 100 using Qualcomm AI Engine direct.
As part of the new offering, the Qualcomm AI Engine direct will now scale across every AI accelerator inside a comprehensive range of Qualcomm Technologies products. Qualcomm AI Engine direct is an AI library that delegates and deploys existing models directly to the AI accelerators on Qualcomm Technologies platforms and will now provide OEMs and developers the ability to develop one feature, then move the same model across different products and tiers.
The Qualcomm AI Stack supports different AI frameworks and popular runtimes including TensorFlow, PyTorch and ONNX, developer libraries and services, system software, tools, and compilers so that any AI feature developed for one device can easily be deployed on others.
The Qualcomm AI Stack portfolio will also include a suite of tools, including Qualcomm AI Model Efficiency Toolkit (AIMET), AI development graphical user interface (GUI), model analyzers to foster enhanced quantization and optimization, and Neural Architecture Search (NAS). At the Snapdragon Tech Summit 2021, Qualcomm announced the Qualcomm Technologies and Google Cloud integrated Google Cloud Vertex AI NAS, which enables OEMs and the ecosystem to create, test, and deploy unique and efficient experiences on the edge.
The Qualcomm Neural Processing SDK remains a key product for OEMs and developers to run their neural networks on a broad range of Qualcomm Technologies’ products and will continue to receive regular updates and support.
The Qualcomm AI Stack is a building block for different SDKs.
The Qualcomm AI Stack will provide Qualcomm Technologies with a baseline to create domain-specific SDKs to deploy across the Connected Intelligent Edge.
For instance, the Qualcomm Neural Processing SDK remains a key product for OEMs and developers to run their neural networks on a broad range of Qualcomm Technologies’ products and will continue to receive regular updates and support.
In sum, Qualcomm’s new AI Stack enables OEM partners and developers to further their investments, build and leverage their AI use cases developed for any Qualcomm business line, and extend it across other business lines or emerging business lines of Qualcomm.