Stable Diffusion 3 API is available

We are pleased to announce the availability of Stable Diffusion 3 and Stable Diffusion 3 Turbo on the Stability AI Developer Platform API.

As shown in the research paper Stable Diffusion 3, this model is as good as or better than state-of-the-art text-to-image generation systems, such as DALL-E 3 and Midjourney v6, in terms of typography and adherence to responsiveness, based on an assessment of human preferences.

The new Multimodal Diffusion Transformer (MMDiT) architecture uses separate sets of weights for image and language representations, which improves text understanding and writing capabilities over previous versions of Stable Diffusion.

While the model is available via API today, we are constantly working to improve it before its open release. In line with our commitment to open generative AI, we intend to make the scales model available for self-hosting with Stability AI membership in the near future.

We are also inviting a limited number of users to participate in an early beta release of our Stable Assistant with support for Stable Diffusion 3. In the limited release, Stable Assistant allows you to create content using advanced image models and the Stability AI language.

Performance

We've partnered with Fireworks AI, the fastest and most reliable API platform on the market, to deliver Stable Diffusion 3 and Stable Diffusion 3 Turbo.

With Fireworks AI, we will also provide an enterprise-grade API solution that provides 99.9% service availability, which is critical for enterprises utilizing our developer platform for their mission-critical generative AI workloads. Our customers will be able to utilize our services for their mission-critical operations with confidence.

Safety

We believe in safe and responsible practices when working with artificial intelligence. This means that we have taken and continue to take reasonable steps to prevent misuse of Stable Diffusion 3 by bad actors. Security begins when our model is trained and continues throughout testing, evaluation, and deployment. By continually collaborating with researchers, experts, and our community, we look forward to continued honest innovation as we refine the model.

At Stable Diffusion 3, we strive to offer adaptable solutions that enable individuals, developers, and businesses to unleash their creativity, in line with our mission to activate humanity's potential.