What you need to know
- Qualcomm AI Hub just launched, giving developers access to over 75 AI models that are optimized to run on Qualcomm processors.
- Those models can be used to perform on-device AI tasks, such as image generation, voice recognition, and real-time translation.
- Running AI models on a device rather than relying on the cloud takes away the need for an internet connection and also improves privacy.
- PCs running Qualcomm Snapdragon X Elite processors are set to ship later this year, and Qualcomm shared the results of a head-to-head comparison between a Snapdragon X Elite-powered PC and one running an Intel Core Ultra CPU.
AI is the biggest buzz word in tech these days, and you shouldn't expect it to go away any time soon. In fact, the latest announcement from Qualcomm shows tech that will bring AI closer to you. The company unveiled the Qualcomm AI Hub, which provides developers with access to over 75 AI models optimized to run on Qualcomm chips. That means that your next smartphone or PC may be able to run powerful AI models locally rather than relying on an internet connection and the cloud.
Qualcomm AI Hub includes some of the biggest names in AI, including image generator Stable Diffusion, speech recognition tool Whisper, and Yolo-v7, which can detect objects in real time. With those models optimized for Qualcomm chips, they should have lower memory utilization and better power efficiency.
With Qualcomm AI Hub, developers should be able to integrate the supported AI models into applications with relatively little effort.
“With Snapdragon 8 Gen 3 for smartphones and Snapdragon X Elite for PCs, we sparked commercialization of on-device AI at scale. Now with the Qualcomm AI Hub, we will empower developers to fully harness the potential of these cutting-edge technologies and create captivating AI-enabled apps,” said Qualcomm Senior Vice President and General Manager of Technology Planning and Edge Solutions Durga Malladi. “The Qualcomm AI Hub provides developers with a comprehensive AI model library to quickly and easily integrate pre-optimized AI models into their applications, leading to faster, more reliable and private user experiences.”
The models are available now through Qualcomm AI Hub, GitHub, and Hugging Face.
"We are thrilled to host Qualcomm Technologies' AI models on Hugging Face," said Clement Delangue, cofounder and CEO, Hugging Face. "These popular AI models, optimized for on-device machine learning and ready to use on Snapdragon and Qualcomm platforms, will enable the next generation of mobile developers and edge AI applications, making AI more accessible and affordable for everyone."
Snapdragon X Elite vs Intel Core Ultra
By the end of the year, both Qualcomm and Intel will have processors optimized for artificial intelligence. But according to Qualcomm, its upcoming Snapdragon X Elite chip beats out its Intel Core Ultra competitor. Qualcomm tested a Snapdragon X Elite-powered laptop against an Intel Core Ultra laptop by having both PCs generate an image in GIMP with the Stable Diffusion plug-in. The Snapdragon X Elite PC finished the task in 7.25 seconds, while the Intel PC took 22.26 seconds.
Of course, this is a specific test, and we don't know all of the parameters used for the head-to-head comparison. The full specs of both PCs are also unknown. At this point, the biggest takeaway from this test is that Qualcomm feels comfortable boasting about better performance than Intel when it comes to generative AI. If times were closer, I think Qualcomm would have chosen a different narrative. We'll have to wait until PCs powered by a Snapdragon X Elite ship later this year to see real-world testing.
It's an exciting year for fans of AI and those that will benefit from its features. Much of Qualcomm's announcement focused on the benefits of running AI models locally on smartphones, but there should be a knock-on effect for PCs. Since Qualcomm is working to optimize AI for its own chips, any device that runs on them should benefit. Qualcomm promised "support for additional operating systems" and that it will add more models to Qualcomm AI Hub over time.
