Back to Tools
Stable Diffusion

Stable Diffusion

Stability

Open generative AI model family from Stability AI for image, video, audio, and 3D creation, available self-hosted, via API, or through partner platforms.

Updated 19d ago
FreeImage Generation⚡ 70

Screenshots

About Stable Diffusion

Stable Diffusion is a family of image-generation models from Stability AI, distinguished from most competitors by the fact that the model weights can be downloaded and run on your own hardware. Stability's model page lists the current lineup: Stable Diffusion 3.5 Large, which Stability describes as the most capable in the family, aimed at professional use at one-megapixel resolution; a Large Turbo variant designed to run faster and produce good images in four steps; and a Medium model that balances quality and customisation and is designed to run on consumer hardware. The company also lists SDXL, a text-to-image model with 3.5 billion parameters, and SDXL Turbo, a distilled version that uses Adversarial Diffusion Distillation to generate in as few as one step.

Because it is a model family and not one app, how you use it depends on where you run it. The page describes four routes: deploy the models in your own environment for deep customisation, integrate them through Stability's API, use them through cloud partners with no extra infrastructure, or work in a web-based creative platform. Alongside generation, the API offers editing tools for images you upload or generate, upscaling tools that increase size and resolution, and tools that transform variations of images or sketches. The Hugging Face model card describes 3.5 Large as a Multimodal Diffusion Transformer text-to-image model, and shows loading it with the diffusers library, after accepting the licence.

Licensing is the practical question for most users. The model card says the model is released under the Stability Community License, and the licence FAQ states that use of the core models is free for everyone unless it is for commercial purposes and the organisation's annual revenue exceeds USD 1 million; below that threshold businesses do not need to pay to use or distribute them. Above it, you contact Stability for commercial terms.

Before commercial use, read the current licence and confirm whether your organisation crosses the revenue threshold. Running the models yourself also depends on your hardware, which the Medium model is designed to accommodate.

Sources

The official pages this summary was checked against. Details such as pricing change; confirm on the vendor site.

Key Information

Pricing
Free
Platforms
Self-hosted, API, Web (via partner platforms)
Category
Image Generation
Developer
Stability
Last Updated
19d ago
Verified
Yes
Website Status
Online
Website
stability.ai

Features

Open-weight text-to-image models: Stable Diffusion 3.5 Large, Large Turbo and Medium
Deploy on your own infrastructure, through an API, via cloud partners or web apps
Image editing, upscaling and variation tools available through Stability AI's API
SDXL Turbo: distilled model for real-time generation in as few as one step
Stability Community License: free below $1M annual revenue, per the licence FAQ
Runs with the Hugging Face diffusers library

Mentioned in Our Blog