stability-ai/stable-diffusion-3.5-large-turbo

A text-to-image model that generates high-resolution images with fine details. It supports various artistic styles and produces diverse outputs from the same prompt, with a focus on fewer inference steps

DocumentationDoc

Idle

A young female mage with waist-length silver hair, speckled with tiny, glittering stars. She wears a deep blue velvet robe embroidered with golden moon phases and constellations. In her hand, she holds a staff made of white crystal, its tip levitating a small, softly glowing nebula. The background is an ancient library with towering bookshelves filled with magical tomes, and light filtering through stained-glass windows. --anime

Your request will cost $0.04 per run.

For $1 you can run this model approximately 25 times.

One more thing:

ExamplesView all

A Viking longship with a dragon head prow, breaking through the waves in a mist-filled Norwegian fjord at dawn. The ship is filled with Viking warriors clad in furs, holding axes and shields, their expressions resolute. The dragon head on the bow is intricately carved, with a fearsome glare. Surrounded by steep cliffs and cascading waterfalls. The color palette is cool and stark, filled with an epic sense of history and adventure, realistic style with attention to historical detail. --ar 21:9

The moment of collision, blending, and explosion of multiple colored inks in water. Deep blues, vibrant magentas, and liquid-gold colors intertwine, forming complex, nebula-like organic shapes. Captured with high-speed photography, extremely sharp details, against a pure black background to emphasize the dynamic and unpredictable beauty of the colors. Abstract art, 4K resolution. --ar 16:9

An elderly artisan repairing a chair outside his workshop in an old Italian alley. He wears reading glasses on his wrinkled face, his expression focused. The afternoon sun casts strong light and shadows across him. Documentary photography style, black and white, high contrast, grainy texture, capturing a candid moment of real life. --style raw --ar 3:2

The busy Shibuya Crossing in Tokyo on a rainy night. Pedestrians hurry by with colorful umbrellas, the ground reflecting the blurred glow of neon signs and car lights. Street photography style, handheld camera perspective with slight motion blur, focused on a solitary figure from behind amidst the crowd. The photo is filled with a sense of urban dynamism and alienation. --ar 16:9

A stylishly dressed model wearing a unique piece of clothing from a Singaporean designer, posing against a backdrop of colorful Peranakan architecture in Joo Chiat. Soft, flattering light highlighting the details and design of the garment. Convey a sense of local culture and contemporary style. --ar 9:16

A male character leaning against a wall covered in neon graffiti. He has sharp black hair, and one of his eyes is covered by a high-tech eyepatch glowing with a faint red light. He wears a black trench coat with futuristic designs, and the seams of a cybernetic arm are visible. He grips the hilt of a high-frequency katana, his gaze cold and intense. The background is a rainy, futuristic city at night, the wet streets reflecting the vibrant lights of advertisements. --ar 9:16 --style raw

A female elf knight wearing lightweight armor crafted from leaves and white metal, with glowing vines wrapped around it. She has long golden hair woven into intricate braids and long, elegant pointed ears. She holds a longbow inlaid with a green gemstone, her expression firm and serene. She stands beneath a giant, bioluminescent ancient tree, surrounded by dancing fireflies. --ar 3:4

README

Stable Diffusion 3.5 Large Turbo is a Multimodal Diffusion Transformer (MMDiT) text-to-image model with Adversarial Diffusion Distillation (ADD) that features improved performance in image quality, typography, complex prompt understanding, and resource-efficiency, with a focus on fewer inference steps. Please note: This model is released under the Stability Community License. Visit Stability AI to learn or contact us for commercial licensing details.

Model Description Developed by: Stability AI Model type: MMDiT text-to-image generative model Model Description: This model generates images based on text prompts. It is an ADD-distilled Multimodal Diffusion Transformer that use three fixed, pretrained text encoders, and with QK-normalization.