Term · Advanced
ControlNet
An add-on that steers image generation using structure like pose, depth, or edges.
ControlNet conditions a diffusion model on structural guides, a pose skeleton, depth map, edge outline, or scribble, so you control composition and shape precisely instead of relying on prompt luck. It's central to the Stable Diffusion ecosystem. Creators use it to reproduce an exact pose, match a layout across images, or enforce consistent framing.
Also called control net, control-net
Related
- Stable DiffusionOpen-source image model you can run locally with deep customization and control.
- Img2ImgGenerating a new image guided by an input image plus a prompt.
- LoRAA small add-on file that tweaks a base AI model toward a specific style, character, or subject.
- Checkpoint ModelA trained base model file that defines an AI image generator's overall style and knowledge.
- Prompt EngineeringThe craft of writing inputs that steer an AI model toward the result you want.