Introducing Stability AI’s latest text-to-image models
Three of Stability AI's newest cutting-edge text-to-image models are now available in Amazon Bedrock, providing high-speed, scalable, AI-powered visual content creation capabilities.
Meet Stability AI
Stability AI is a world-leading open source generative AI company delivering breakthrough AI models with minimal resource requirements in imaging, language, code and audio.
Benefits
Stride Learning Cultivates Student Creativity Using Stability AI on Amazon Bedrock
Stride Learning's Legend Library is revolutionizing reading comprehension for K-4 students. Their AI-driven educational app lets kids bring their stories to life with stunning illustrations created by Stability AI's Stable Diffusion, ensuring accurate and safe content.
Revolutionize Creative Workflows with SD3
Stable Diffusion 3 has the potential to revolutionize the creative workflow in advertising and marketing. By synergizing LLM-generated campaign ideas with advanced image generation capabilities, agencies can rapidly produce high-quality, tailored visual assets that resonate with their target audience.
Use cases
Model Features
Photorealism
Stable Image Ultra delivers images with exceptional detail, color, and lighting, enabling photorealistic outputs as well as high-quality outputs in flexible styles.
This image was generated using Stable Image Ultra with the prompt: An ornate, Victorian-era key lying on a weathered, wooden surface, with intricate, steampunk-inspired gears and mechanisms visible within its transparent, glass shaft
Prompt Understanding
Stability AI models are able to comprehend long and complex prompts involving spatial reasoning, compositional elements, actions, and styles.
This image was generated using Stable Image Ultra with the prompt: a dramatic studio headshot of a male subject with strong, directional lighting. Use a 100mm lens to capture a tight headshot with excellent background compression. Set the aperture to f/4 to keep the entire face in sharp focus while maintaining a pleasing depth of field. Use a shutter speed of 1/125s to sync with studio strobes and ensure proper exposure. Focus on the subject's eyes to draw the viewer into the portrait.
Typography
Stable Image Ultra achieves unprecedented text quality with fewer errors in spelling, kerning, letter forming, and spacing. Here, SD3 Ultra is able to generate specific text, objects, and lighting conditions accurately.
This image was generated using Stable Image Ultra with the prompt: photo of blossom flowers falling from a tree, making the word "Spring", during sunset
High Quality Illustrations
SD3 Large can generate high quality illustrations, paintings, and more ensuring accurate and engaging visuals for many different publications.
This image was generated using SD3 Large with the prompt: oil painting, a woman sitting in a field watching a kite fly in the sky, traditional media, oil paint, digitalized, highly detailed, concept art, intricate, professional composition
Product Rendering
Stability AI’s models can be used to generate high-quality concept art and product renderings or creating visually stunning images for billboards and print ads.
This image was generated using SD3 Large with the prompt: professional 3d render of a white and teal sneaker, floating in center, hovering, floating, high quality, photorealistic
Model versions
Stable Image Ultra
Stable Image Ultra offers enhanced creativity and photorealism, rendering exceptional visuals with highly detailed 3D imagery that include fine details like lighting, textures, and hands. The model boasts proficiency in generating images with multiple subjects, making it ideal for creating complex scenes..
Languages: English
Fine Tuning Supported: No
Supported use cases: Media and Entertainment, Game Development, Retail, Publishing, Education and Training, Marketing/Advertising Agencies
Stable Diffusion 3 Large (SD3 Large)
This model offers greatly improved performance in multi-subject prompts, image quality, and spelling abilities. SD3 Large introduces a new state-of-the-art architecture for image generation, comprising an 8 billion parameter ensemble pipeline that delivers unprecedented quality, ease of use, and the ability to create complex compositions with basic natural language prompting.
Languages: English
Fine Tuning Supported: No
Supported use cases: Media and Entertainment, Game Development, Retail, Publishing, Education and Training, Marketing/Advertising Agencies
Stable Image Core
Create images quickly and affordably with this 2.6 billion parameter model, the flagship service that requires no prompt engineering to get high quality images in diverse styles. Capabilities include improved scene layout, including object placement, versatility and readability across different sizes and applications.
Languages: English
Fine Tuning Supported: No
Supported use cases: Media and Entertainment, Game Development, Retail, Publishing, Education and Training, Marketing/Advertising Agencies
Stable Diffusion XL 1.0
SDXL is the previous model version, Stable Image Core is the next generation.
Languages: English
Fine-tuning supported: No
Supported use cases: Advertising and marketing, media and entertainment, gaming and metaverse.
Did you find what you were looking for today?
Let us know so we can improve the quality of the content on our pages.