Stability AI Models

Stability AI is the lab behind Stable Diffusion, the open-weight model family that made high-quality image generation broadly accessible, and it is the second-largest partner catalogue on Segmind. The collection spans the full lineage: Stable Diffusion 3.5 and SD3 for current-generation text-to-image quality, SDXL and SDXL Turbo for the widely-adopted 1024px workhorse and its distilled fast variant, and the SD 1.5 and 2.1 pipelines that remain the base for an enormous ecosystem of community LoRAs and fine-tunes. Alongside the base models sit the control and editing pipelines that make Stable Diffusion practical for production work: ControlNet variants for pose, depth, canny-edge and scribble guidance, inpainting and outpainting models for targeted region edits, image-to-image for style-preserving transformation, and Stable Video Diffusion for animating stills. Stability's open-weight approach is what makes this collection distinctive: the models are transparent, extensively documented, and backed by the largest fine-tuning community in generative imaging, so a technique you find in a paper or a community post is usually reproducible here. On Segmind, every Stability model is a pay-per-use API endpoint with no GPU to provision. Chain them with upscalers, face restoration and background removal in Segmind Workflows to build complete automated image production pipelines.