Stability AI Unveils Advanced Text-to-Image Model: Stable Diffusion 3 Medium
Stability AI Introduces Stable Diffusion 3 Medium
Stability AI has announced the launch of Stable Diffusion 3 Medium, the latest and most advanced text-to-image AI model in its Stable Diffusion 3 series. This new model aims to set a new standard for AI-generated art by providing enhanced performance and accessibility, according to Stability AI.
Collaborations Enhancing Performance
In developing Stable Diffusion 3 Medium, Stability AI collaborated with tech giants NVIDIA and AMD to optimize the model's performance. Utilizing NVIDIA® RTX™ GPUs and TensorRT™, the TensorRT-optimized versions of Stable Diffusion models deliver a 50% increase in performance.
Additionally, AMD has optimized inference for SD3 Medium across various AMD devices, including their latest APUs, consumer GPUs, and MI-300X Enterprise GPUs. These collaborations ensure that users experience best-in-class performance when using the new model.
Commitment to Open and Accessible AI
Stability AI maintains its commitment to open generative AI with the release of Stable Diffusion 3 Medium under the Stability Non-Commercial Research Community License. Professional artists, designers, developers, and AI enthusiasts are encouraged to use the new Creator License for commercial purposes. For large-scale commercial use, enterprises can contact Stability AI for licensing details.
Available via API and Applications
Stable Diffusion 3 Medium is accessible through Stability AI's API. Other versions, such as SD3 Large and SD3 Ultra, are also available for trial through the Stable Assistant chatbot and Discord's Stable Artisan. Users can start with a three-day free trial to explore the capabilities of these models.
Safety Measures and Future Plans
Stability AI places a strong emphasis on safe and responsible AI practices. The company has taken extensive measures to prevent the misuse of Stable Diffusion 3 Medium, including rigorous internal and external testing and the implementation of numerous safeguards.
Looking forward, Stability AI plans to continuously improve Stable Diffusion 3 Medium based on user feedback, expand its features, and enhance its performance. The goal is to make Stable Diffusion 3 Medium an indispensable tool for both professionals and hobbyists in the realm of AI-generated art.
For more information on Stability AI's approach to safety, visit their Stable Safety page. To stay updated on future developments, follow Stability AI on Twitter, Instagram, LinkedIn, and join their Discord Community.