The Qualcomm Neural Processing SDK provides tools for model conversion, execution, supports convolutional neural networks, custom layers and more. The SDK is designed to help developers run one or more neural network models trained in TensorFlow, PyTorch, Keras and ONNX on Qualcomm® platforms, whether that is the CPU, GPU or Qualcomm® Hexagon™ NPU.

The Qualcomm® AI Engine Direct SDK provides lower-level, unified APIs for AI development. Developers can get one step closer to the silicon and improve the performance of their AI models on Qualcomm® AI accelerators: Qualcomm® Kryo™ CPU, Qualcomm® Adreno™ GPU and Qualcomm® Hexagon™ NPU. They can use the SDK to target a specific accelerator, or they can delegate workloads from TensorFlow Lite or the ONNX runtime to directly access our Hexagon processor.

The Qualcomm IM SDK provides developers with intuitive APIs and a wide range of plugins in both the multimedia and machine learning domains. The comprehensive suite of plugins includes video encode/decode, Camera ISP, GPU, display, audio DSP (aDSP), Qualcomm® Hexagon™ NPU, and additional AI/ML accelerators. These plugins are meticulously optimized to deliver exceptional performance on Qualcomm® hardware, so that the end applications run easily and efficiently.

The Qualcomm TensorFlow Lite SDK provides the TensorFlow Lite framework for on-device artificial intelligence (AI) inferencing, which facilitates application developers to develop or run suitable AI applications.

Experience the power of the Qualcomm Intelligent Multimedia Software Development Kit (IM SDK) and unlock the full potential of your IoT applications with advanced multimedia and AI capabilities.