Qualcomm demos fastest local AI image generation with Stable Diffusion on mobile


Are you ready to experience the fastest-ever deployment of AI image generator Stable Diffusion on a smartphone? Qualcomm is leading the way in mobile AI technology, with their flagship Snapdragon 8 Gen 2 chipset powering a demo of version 1.5 of Stable Diffusion that can generate a 512 x 512 pixel image in under 15 seconds. If you’re curious about the implications of this breakthrough, read on to learn more about what Qualcomm’s AI chops mean for mobile technology.

What is Stable Diffusion?

Stable Diffusion is a staple in AI image generation, and it typically takes a lot of computing power to run. Most apps offering such services on mobile do all their processing in the cloud rather than on the device, and even generating an image in this way on a decent laptop will take minutes. Qualcomm’s engineers did all sorts of custom optimizations on the software side to get Stable Diffusion running optimally, and this has enabled them to generate images on a smartphone in seconds.

Qualcomm’s Speed Record

Qualcomm claims that their demo is a speed record, and we’ve no reason to doubt it. They also say that it’s the first time Stable Diffusion has ever run locally on Android, though after a bit of digging around, we found a blog post from developer Ivon Huang that shows how they got Stable Diffusion running on a Sony Xperia 5 II powered by a Qualcomm Snapdragon 865 and 8GB of RAM. Huang also noted that generating a 512 x 512 image with this setup took an hour, so Qualcomm still wins points for speed.

Comparing with iOS

Apple released the optimizations needed to get Stable Diffusion running locally on its machine learning framework Core ML in December, so to test the system today, we got Stable Diffusion 1.5 running on an iPhone 13 via the Draw Things app with Core ML acceleration. With this setup, it took about a minute to generate a 512 x 512 image, so again, Qualcomm wins on speed.

The Implications of Qualcomm’s Breakthrough

Getting big AI models running locally on mobile devices offers all sorts of advantages over relying on cloud compute. There’s convenience (you don’t need a mobile connection), cost (developers won’t charge users when the server bills come due), and privacy (running locally means you don’t send data to someone else’s computer). It’s the productization of AI, and it’s happening fast.

Qualcomm is leading the way in mobile AI technology, with their flagship Snapdragon 8 Gen 2 chipset powering a demo of version 1.5 of Stable Diffusion that can generate a 512 x 512 pixel image in under 15 seconds. This breakthrough showcases the potential of AI on mobile devices and the advantages it can bring to users in terms of convenience, cost and privacy. If you’re curious about the implications of Qualcomm’s AI chops for mobile technology, this blog post has the details you need to know.

Leave a comment

Your email address will not be published. Required fields are marked *