
In an official post, social media platform Snapchat announced the upcoming launch of a new AI text-to-image generation model. According to Snapchat, this model is so advanced that it will allow Snapchat users to generate high-quality images on mobile devices in "just seconds." The feature is expected to roll out in the coming months.
The company states that the feature is infused with an advanced AI-powered diffusion model that is designed to be compact yet incredibly fast. Notably, the new AI text-to-image model can generate high-quality images in approximately 1.4 seconds on an iPhone 16 Pro Max. The new diffusion model works entirely on the device, which helps reduce the computation costs that would have been incurred if the model relied on large servers for its operation.
Snap claims it achieved this efficiency by adopting innovative training techniques, allowing its model to learn from larger AI systems while maintaining its small size. Snap has plans to integrate this new AI text-to-image model into Snapchat's popular features such as AI Snaps, AI Bitmoji Backgrounds, and more.
Additionally, this in-house AI technology will allow Snap to offer its community of users high-quality AI tools at a low operating cost. "Snap has a long history of research excellence in model optimization and efficiency. We are inspired by the industry innovation that is making AI tools more efficient, affordable, and accessible and we look forward to continuing to contribute to the rapid pace of innovation, particularly for mobile-first experiences," Snapchat stated.
While the company did not reveal any exact timeline for the launch of this new AI feature, except to say it's arriving in the coming months, it did reveal that this new feature is a continuation of its long-term investment in cutting-edge AI and ML technologies.
1 Comment - Add comment