Can Stable Diffusion's generative AI work on a smartphone? Qualcomm says - Yes it is possible

Check out all the Smart Security Summit on-demand sessions here.

The world's most advanced generative AI models, like Stable Diffusion, typically only work in the cloud.

But what if that same model could work on a smartphone in your pocket? This is the challenge that Qualcomm engineers have taken up. In research published today, Qualcomm found that by using a combination of software techniques and hardware optimization, it was able to reduce stable streaming so that it could run inference models on popular Android smartphones.

Stable Diffusion is developed by startup Stability AI and is one of the most popular generative AI models for imaging currently in use, often competing with OpenAI's DALL-E.

To be clear, the technology needed to train generative AI models is massive and won't run on a smartphone. What Qualcomm has rather worked on is the inference side, that is to say the “generative” part, which makes it possible to create a new image from the pre-trained model. To date, users have been able to generate Stable Streaming-based images on their phones using an indirect approach, where a mobile app or browser accesses a cloud service that generates the image. What Qualcomm is now demonstrating is the ability to generate stable streaming generative AI images directly on an Android smartphone, without needing to call in the cloud to do the heavy lifting.

On-Demand Smart Security Summit

Learn about the essential role of AI and ML in cybersecurity and industry-specific case studies. Watch the on-demand sessions today.

>>Follow VentureBeat's ongoing generative AI coverage