Neura Market
Stable Diffusion Public Release logo

Stable Diffusion Public Release

Free

Announcement of the public release of Stable Diffusion, an AI-based image generation model trained on a broad internet scrape and licensed under a Creative ML OpenRAIL-M license. Stable Diffusion blog, 22 August, 2022.

FreeFree tier
Inputs: textOutputs: image
Type
Open Source
Company
Stability AI
Visit Website GitHub (40 stars)

About Stable Diffusion Public Release

Stable Diffusion is an open-source AI model for generating images from text descriptions. It was publicly released on August 22, 2022 by Stability AI in collaboration with HuggingFace and CoreWeave. The model is licensed under the Creative ML OpenRAIL-M license, which permits both commercial and non-commercial use. A built-in AI-based safety classifier is included by default to filter undesirable outputs. The recommended model weights are v1.4, requiring approximately 6.9 GB of VRAM, and NVIDIA GPUs are recommended. Future optimizations for AMD and Apple Silicon are planned. The release includes the model weights, code, and a HuggingFace diffusers notebook for easy development.

Key Features

Text-to-image generation from natural language prompts
Open-source model weights available (v1.4 recommended)
Creative ML OpenRAIL-M license (commercial and non-commercial use)
Built-in AI-based safety classifier to filter outputs
Optimized for NVIDIA GPUs (6.9 GB VRAM required)
Integration with HuggingFace diffusers library for easy experimentation

Pros & Cons

Pros
  • Free and open-source with permissive license
  • Strong community support and active development
  • Safety classifier included to reduce harmful outputs
  • Available for both commercial and non-commercial projects
Cons
  • Requires 6.9 GB VRAM, limiting accessibility on older hardware
  • Best performance on NVIDIA GPUs; AMD and Apple Silicon support was not ready at launch
  • Potential ethical concerns around generated content (mitigated by safety classifier)

Best For

Creative design and digital art generationContent creation for marketing and advertisingRapid prototyping and concept visualizationResearch and experimentation in generative AI

FAQ

What license is Stable Diffusion released under?
It is released under the Creative ML OpenRAIL-M license, a permissive license that allows commercial and non-commercial use, with ethical and legal responsibilities.
What are the system requirements?
The model requires 6.9 GB of VRAM and is optimized for NVIDIA GPUs at launch. Optimizations for AMD and Macbook M1/M2 were planned.
Does the model include safety features?
Yes, a built-in AI-based safety classifier is included by default to understand concepts and remove outputs that may be undesirable.
Where can I find the model weights and code?
Model weights, card, and code are available on the Stability AI website. An optimized notebook using the HuggingFace diffusers library is also provided.
Is there a commercial product based on Stable Diffusion?
Yes, DreamStudio is a public demonstration space and commercial offering for faster generation and additional control.