This initiative aligns with broader industry trends, as many tech companies, including Meta, are investing heavily in their own AI technologies.
The company has a strong track record in research focused on model optimization and efficiency, as highlighted in a recent blog post.
Notably, this new diffusion model operates entirely on-device, significantly reducing computational costs compared to traditional server-dependent models.
Snap asserts that the AI model can generate visually stunning results by leveraging representations from large-scale diffusion models, optimized for efficiency.
While the full range of features powered by this new AI model has yet to be revealed, there is considerable anticipation for upcoming developments.
The launch of this model marks a significant shift for Snap, moving away from reliance on AI tools from OpenAI and Google to developing an independent AI solution.
This announcement follows a series of AI-powered features introduced by Snap, particularly since the release of Google's Gemini.
The model is capable of producing high-resolution images in about 1.4 seconds on an iPhone 16 Pro Max, although users with less powerful devices may experience longer generation times.
Currently, the AI text-to-image model is still in production, with expectations for its release in the near future.
Snap has unveiled a new AI text-to-image model designed specifically for mobile devices, which is set to enhance the platform's features in the coming months.
This innovative technology will be integrated into various features, including AI Snaps and AI Bitmoji Backgrounds, further enriching user experience.
By creating this in-house technology, Snap aims to deliver high-quality AI tools while simultaneously reducing operational costs.
