Snapchat just introduced a new text-to-image diffusion model for generative AI. The new approach provides a faster model for building visuals based on text queries. Most generative AI tools currently take about 30 seconds to generate such images. But Snap says its new system can produce similar images in less than two seconds. It is undoubtedly an interesting development in the broader context of the generative AI process.
As explained by Snap:
“SnapFusion reduces the model’s run time from text input to image generation on mobile to less than two seconds – the fastest time published by the academic community to date. Snap Research achieved this breakthrough by optimizing the network architecture and noise reduction process, making it incredibly efficient while maintaining image quality. So it is now possible to run the model to generate images from text prompts, and get sharp, clear images back on mobile devices in seconds, rather than minutes or hours, as other research shows.”
Snapchat Research introduces a new text-to-image diffusion model for generative AI
Also check: How to Remove Snapchat Filter from Saved Photo?
Snapchat also explained some examples of the images produced by the SnapFusion process. It’s much like the same type of generative AI graphics you get from any other app. The result time is quite faster, which Snap says could have a number of benefits.
Aside from that, Snap also notes that the new process could facilitate improved privacy. It also limits data sharing with third parties, while also reducing processing costs for developers.
It’s worth noting here that most of his experiments were conducted on an iPhone Pro 14, which, in Snap’s own words, “has more computing power than many other phones.” It is uncertain whether other phones will also be faster or not.
Also see: How To Fix Snapchat Cannot Load Pictures Problem?