Snap has announced the launch of a new research AI image generation model that will be available on mobile devices. This model is capable of creating high-quality images in just 1.4 seconds on the iPhone 16 Pro Max. It operates entirely on-device, reducing computational costs compared to models that require powerful servers.
The company emphasizes that the new diffusion model can produce “stunning” visual results by transferring rich representations from large diffusion models to this smaller and more efficient model. Snap plans to implement this technology in production over the coming months to support features such as AI Snaps and AI Bitmoji Backgrounds.
Thanks to the introduction of this in-house technology, Snap will be able to offer its community high-quality AI tools at lower operational costs. The company notes that it has extensive experience in optimizing and improving the efficiency of models. It continues to invest significantly in the development of AI and machine learning technologies.
Previously, Snap used AI tools from companies such as OpenAI and Google to implement some of its features. However, the company has now introduced its own model, which will allow it to offer users integrated capabilities.
