Qualcomm Releases IMSDK 2.0: Bridging Generative AI and Multimedia, Accelerating Edge AI Application Development

Beating AI News – Qualcomm has announced the release of Qualcomm Intelligent Multimedia SDK (IMSDK) 2.0, a unified AI and multimedia development framework designed for Qualcomm Dragonwing and Linux edge platforms. The SDK aims to simplify the integration of features such as cameras, audio and video, sensors, AI inference, and cloud connectivity, thereby accelerating the deployment of products like smart cameras, robots, drones, and industrial AI solutions. IMSDK 2.0 introduces new features including Python and C++ Pipeline APIs, an application builder, generative AI inference, containerized microservices, AI programming assistant capabilities, and "documentation as code." Developers can switch AI models without restructuring the overall multimedia pipeline by leveraging various inference runtimes such as QAIRT, ONNX Runtime, and TFLite. Qualcomm stated that IMSDK 2.0 is built on the GStreamer architecture, incorporating hardware-accelerated plugins, zero-copy data transfer, and multi-stream, multi-model pipeline capabilities to support applications like visual AI, LLM/VLM, audio AI, and text-to-image generation. Its containerized microservices cover inference, analytics, and platform services, and support enterprise-grade connectivity options such as Kafka, MQTT, AWS IoT, and Azure IoT. Additionally, IMSDK 2.0 introduces AI programming assistant capabilities, enabling developers to configure pipelines, integrate models, compile, debug, and deploy through natural language assistance. Qualcomm noted that this SDK has already been adopted by companies such as Samsung, Amazon, and Bose for related product development. [Original Link]