Stable Diffusion 2.1 available at 768x768 resolution with improved NSFW filtering. Stable UnCLIP 2.1 introduced at 768x768 resolution with CLIP ViT-L/H variants. Version 2.0 includes x4 upscaling and depth-guided models
Graph/nodes interface for designing and executing stable diffusion pipelines. Supports multiple diffusion models including SD1.x, SD3, Stable Cascade, and Stable Video. Includes asynchronous queue system and smart memory management. Works offline without downloading anything
Stable Diffusion generates photo-realistic images from text input. Developed by Robin Rombach and Patrick Esser. Uses CLIP ViT-L/14 text encoder as pretrained model. Available in English language only
Web interface for Stable Diffusion using Gradio library. Supports txt2img and img2img modes with one-click installation. Includes outpainting, inpainting, color sketch, and prompt matrix. Offers multiple upscaling options including ESRGAN and LDSR
AI-powered tool generates images directly from text input. Uses Stable Diffusion model released in 2022. Provides precise and realistic image generation. Automates image synthesis process. Offers user-friendly web-based interface
Free AI image generator creates high-quality images from text descriptions. Users get 5 free credits without login. Generates images in various styles including anime, fantasy, and cyberpunk. Images are 1024x1024 PNG format with 20-40 second generation time