Stable Diffusion is an open-source, deep-learning model designed primarily to generate high-quality images from text descriptions (text-to-image). Released in August 2022 by Stability AI in collaboration with researchers from LMU Munich and Runway, it has become a cornerstone of the generative AI boom due to its accessibility and high level of customization.

Key Features
1. Open-Source & Free: Unlike proprietary models like DALL-E or Midjourney, Stable Diffusion’s code and model weights are publicly available for download and modification.
2. Local Execution: It is efficient enough to run on consumer-grade hardware (PCs with a dedicated GPU), providing users with full privacy and no subscription fees.
3. Versatile Capabilities: Beyond creating images from scratch, it can:
– Image-to-Image: Modify existing images based on a new prompt.
– Inpainting & Outpainting: Replace parts of an image or extend it beyond its original borders.
– Video & Animation: Generate short clips using community extensions like Deforum.