AI image generation glossary

Every term here changes something you can see in your output. No maths, no hand-waving, and an honest note where Dreamify does not expose the control being described.

Fundamentals

What is Stable Diffusion?

Stable Diffusion is an open text-to-image model that generates pictures by starting from random noise and repeatedly removing it, guided by your prompt.

What is a diffusion model?

A diffusion model generates data by learning to reverse a noising process, turning random static into an image one denoising step at a time.

What is latent space?

Latent space is a compressed numerical representation of an image, where diffusion models do all their work before decoding back to pixels.

What is a negative prompt?

A negative prompt is a second description of what you do not want, which the model actively steers away from during generation.

What is a seed in AI image generation?

The seed is the number that determines the random starting noise, so the same seed and prompt reproduce the same image every time.

What is a scheduler (sampler)?

The scheduler, or sampler, is the algorithm that decides how large each denoising step is and how the model moves from noise toward the final image.

What are inference steps?

Inference steps are the number of denoising passes the model runs, trading generation time against how far the image is refined.

What is CFG scale (guidance scale)?

CFG scale, or guidance scale, controls how strongly the model is pushed toward your prompt and away from the negative prompt at each step.

How to write AI image prompts

Good image prompts name the subject, the setting, the lighting, the medium and the framing, in roughly that order, using concrete nouns rather than praise.

What is a LoRA?

A LoRA is a small add-on file that adjusts a model's behaviour toward a specific style, character or concept without retraining the whole model.

What is AI upscaling?

AI upscaling enlarges an image by generating plausible new detail, rather than simply stretching the pixels that are already there.

What is inpainting?

Inpainting regenerates only a masked region of an existing image, using the surrounding pixels as context so the new content blends in.

Put it into practice

Enable advanced features in the generator to see the negative prompt, scheduler, inference steps and seed for yourself.

Start generating

Next, browse the use case guides, the art styles, or the models.