Qualcomm has been making waves at the Mobile World Congress 2024 by showcasing its impressive portfolio of AI-on-phone tricks facilitated by the Snapdragon series silicon for Android phones. Building on the company’s previous demonstrations of AI capabilities for the Snapdragon 8 Gen 3 flagship, Qualcomm has now added more AI superpowers to its arsenal. The latest updates include the ability to run a Large Language and Vision Assistant (LLaVa) on a smartphone, which enables the processing of images and text inputs, effectively turning a smartphone into a chatbot with Google Lens abilities.
The LLaVa assistant is based on a large multimodal model (LMM) that can process over 7 billion parameters, allowing it to handle complex tasks such as image recognition, object detection, and answering follow-on queries. This on-device processing capability is a significant departure from cloud-based AI solutions, which can be slower and more prone to privacy risks. Qualcomm’s approach ensures faster performance, enhanced privacy, and reliability, making it an attractive option for users.
The company is also showcasing its creative domain of image generation and manipulation by introducing LoRA-driven image generation. LoRA, short for Low-Rank Adaptation, is a technique developed by Microsoft that dramatically reduces the model weight by only focusing on specific segments of the model and reducing the number of parameters for training purposes. This results in faster processing times, reduced memory requirements, and easier adaptability, making it an ideal solution for smartphones.
Additionally, Qualcomm is demonstrating other AI tricks, including the ability to expand the canvas of an image using generative AI fill and AI-powered video generation. The latter is particularly ambitious, as it aims to bring advanced AI capabilities typically reserved for data centers and cloud servers to smartphones. Qualcomm’s implementation of these technologies has the potential to revolutionize the way users interact with their devices, enabling more intuitive and personalized experiences.
Qualcomm’s focus on on-device processing and AI capabilities has significant implications for the future of mobile technology. With LLaVa and LoRA-driven image generation, smartphones are being transformed into powerful AI-driven devices that can handle complex tasks independently, without the need for cloud connectivity. This shift towards on-device processing will likely have a profound impact on the way users interact with their devices, enabling faster and more seamless experiences, while also reducing the risk of privacy breaches.
Qualcomm’s showcase at Mobile World Congress 2024 is a testament to the company’s commitment to pushing the boundaries of what is possible with AI-on-phone capabilities. The announcement of LLaVa and LoRA-driven image generation is a significant milestone in the development of AI-powered smartphones, and has the potential to revolutionize the way users interact with their devices. As the technology continues to evolve, it will be exciting to see how Qualcomm’s innovations shape the future of mobile technology.