Snap has announced the development of a new AI text-to-image model designed for mobile devices. This model will soon enhance several Snapchat features. According to the company, the model can generate high-resolution images in approximately 1.4 seconds on an iPhone 16 Pro Max.

The innovative diffusion model operates entirely on the device, which lowers computational costs compared to models that depend on large servers. Snap highlights that the model delivers impressive visual results by adapting rich representations from large-scale diffusion models to this more efficient version.

Snap intends to integrate this technology into its app in the coming months, enhancing AI Snaps, AI Bitmoji Backgrounds, and more. By utilizing in-house technology, Snap aims to provide its community with high-quality AI tools at a reduced operational cost.

"Snap has a long history of research excellence in model optimization and efficiency," the company mentioned in a blog post. "We are inspired by the industry's advancements in making AI tools more efficient, affordable, and accessible, and we look forward to contributing to the rapid pace of innovation, particularly for mobile-first experiences."

While the company did not disclose further specifics about the model, it emphasized that this development is a continuation of its long-term investment in AI and machine learning technologies.

In the past, Snap has leveraged AI tools from companies like OpenAI and Google to enhance its features. However, with the introduction of its proprietary AI model, Snap is now positioned to offer unique capabilities to its users. This move aligns with the broader industry trend of tech companies, including Meta, investing in AI development.

Snap is scheduled to release its Q4 2024 earnings report on Tuesday.

Image by TechCrunch