Stable Diffusion
Free tierOpen-source image generation model — run locally, fine-tune on your own images, full control.
Free·Technical·Powered by Stability AI·API available·Open source
Key strengths
fully open-sourcelocal executionLoRA fine-tuninghuge model ecosystem
Completely free
London, UK
Founded 2022
Self-hostable
No ratings yet
Stable Diffusion is a latent diffusion model. SDXL and SD 3.5 are the current generations. Run locally via ComfyUI, Automatic1111 WebUI, or InvokeAI. Fine-tune with LoRA (Low-Rank Adaptation) using 15–30 images. The Stability AI API offers cloud inference. The model weights are freely downloadable from Hugging Face for commercial use (check the license per model version).