Snap claims its AI can produce visually stunning results by optimizing representations from large-scale diffusion models.

Snap has unveiled an AI text-to-image model designed for mobile devices, set to enhance Snapchat's features in the coming months.

By developing its own AI model, Snap aims to offer users advanced creative tools while decreasing dependence on external providers such as OpenAI and Google.

The company plans to integrate this technology into features like AI Snaps and AI Bitmoji Backgrounds.

The model is capable of generating high-resolution images in about 1.4 seconds on an iPhone 16 Pro Max, emphasizing both speed and efficiency.

Users with less powerful devices may experience longer image generation times, although the base time remains relatively quick.

Operating on-device allows Snap to reduce costs while maintaining impressive visual quality, in contrast to traditional cloud-based systems.

This in-house technology development is intended to provide high-quality AI tools while lowering operational costs.

The introduction of this model signals increased competition in mobile AI technology, as companies strive to deliver faster and more efficient features.

This initiative aligns with a broader industry trend, as many tech companies, including Meta, are investing heavily in their own AI technologies.

While the full range of features powered by the new AI model has yet to be disclosed, anticipation for further developments remains high.

This announcement follows a series of AI-powered features introduced by Snap since the launch of Google's Gemini.