Announcing the Open Release of Stable Diffusion 3 Medium, Our Most Sophisticated Image Generation Model to Date
Key Takeaways
Stable Diffusion 3 Medium is Stability AI’s most advanced text-to-image open model yet.
The small size of this model makes it perfect for running on consumer PCs and laptops as well as enterprise-tier GPUs. It is suitably sized to become the next standard in text-to-image models.
The weights are now available under an open non-commercial license and a low-cost Creator License. For large-scale commercial use, please contact us for licensing details.
To try Stable Diffusion 3 models, try using the API on the Stability Platform, sign up for a free three-day trial on Stable Assistant, and try Stable Artisan via Discord.
We are excited to announce the launch of Stable Diffusion 3 Medium, the latest and most advanced text-to-image AI model in our Stable Diffusion 3 series. Released today, Stable Diffusion 3 Medium represents a major milestone in the evolution of generative AI, continuing our commitment to democratising this powerful technology.
What Makes SD3 Medium Stand Out?
SD3 Medium is a 2 billion parameter SD3 model that offers some notable features:
Overall Quality and Photorealism: Delivers images with exceptional detail, color, and lighting, enabling photorealistic outputs as well as high-quality outputs in flexible styles. Success in addressing common pitfalls of other models, such as realism in hands and faces, is achieved through innovations such as the 16-channel VAE.
Prompt Understanding: Comprehends long and complex prompts involving spatial reasoning, compositional elements, actions, and styles. By utilizing all three text encoders or a combination, users can trade off performance for efficiency.
Typography: Achieves unprecedented text quality with fewer errors in spelling, kerning, letter forming, and spacing by leveraging our Diffusion Transformer architecture.
Resource-efficient: Ideal for running on standard consumer GPUs without performance degradation, thanks to its low VRAM footprint.
Fine-Tuning: Capable of absorbing nuanced details from small datasets, making it perfect for customisation.
Our collaboration with NVIDIA
We collaborated with NVIDIA to enhance the performance of all Stable Diffusion models, including Stable Diffusion 3 Medium, by leveraging NVIDIA® RTX™ GPUs and TensorRT™. The TensorRT- optimised versions will provide best-in-class performance, yielding a 50% increase in performance.
Download the TensorRT-optimised version of Stable Diffusion 3 Medium.
Our collaboration with AMD
AMD has optimized inference for SD3 Medium for various AMD devices including AMD’s latest APUs, consumer GPUs and MI-300X Enterprise GPUs.
Open and Accessible
Our commitment to open generative AI remains unwavering. Stable Diffusion 3 Medium is released under the Stability Non-Commercial Research Community License. We encourage professional artists, designers, developers, and AI enthusiasts to use our new Creator License for commercial purposes. For large-scale commercial use, please contact us for licensing details.
Try Stable Diffusion 3 via our API and Applications
Alongside the open release, Stable Diffusion 3 Medium is available on our API powered by Fireworks Al. Other versions of Stable Diffusion 3 such as the SD3 Large model and SD3 Ultra are also available to try on our friendly chatbot, Stable Assistant and on Discord via Stable Artisan. Get started with a three-day free trial.
How to Get Started
Download the weights of Stable Diffusion 3 Medium
For Commercial Inquiries: Contact us for licensing details.
Start a three-day trial of our image services on Stable Assistant and Stable Artisan.
Learn more about the model in our detailed FAQs.
Safety
We believe in safe, responsible AI practices. This means we have taken and continue to take reasonable steps to prevent the misuse of Stable Diffusion 3 Medium by bad actors. Safety starts when we begin training our model and continues throughout testing, evaluation, and deployment. We have conducted extensive internal and external testing of this model and have developed and implemented numerous safeguards to prevent harms.
By continually collaborating with researchers, experts, and our community, we expect to innovate further with integrity as we continue to improve the model. For more information about our approach to Safety please visit our Stable Safety page.
Licensing
Stable Diffusion 3 Medium is released under the Stability Non-Commercial Research Community License.
We have introduced the new Creator License to encourage the community to leverage Stable Diffusion 3 while supporting Stability in the mission to keep AI open and accessible. We encourage professional artists, designers, developers, and AI enthusiasts to use the Creator License to start building with Stable Diffusion.
Large-scale commercial users and enterprises are requested to contact us and obtain an Enterprise License. This will help us ensure that businesses can leverage the full potential of our model while adhering to our usage guidelines.
Future Plans
We plan to continuously improve Stable Diffusion 3 Medium based on user feedback, expand its features, and enhance its performance. Our goal is to set a new standard for creativity in AI-generated art and make Stable Diffusion 3 Medium a vital tool for professionals and hobbyists alike.
We are excited to see what you create with the new model and look forward to your feedback. Together, we can shape the future of generative AI.
To stay updated on our progress follow us on Twitter, Instagram, LinkedIn, and join our Discord Community.
