Snap has introduced a mobile AI text-to-image model that will enhance several features of Snapchat in the near future. On Tuesday, the company announced that this model can generate high-resolution images in approximately 1.4 seconds on an iPhone 16 Pro Max.
This innovative diffusion model operates entirely on the device itself, significantly decreasing computational expenses compared to models dependent on extensive server resources. Snap highlights that this model is capable of producing “breathtaking” visuals by effectively “transferring rich representations from expansive diffusion models to this more compact and efficient version.”


