Glossary

Seed in AI art

What is Seed in AI art?

The seed is a starting number for an image model's random number generator. Diffusion models begin with random image noise from which a picture gradually emerges; the seed determines what this initial noise looks like. With the same seed, prompt and settings, an image can largely be reproduced in the same model. This helps when changing just one detail in the prompt or recreating a successful image variant later.

Topics

  • image generation
  • reproducibility
  • randomness

Try KI.ordo

Agents, flows and templates in one workspace – free 24-hour demo.

Start demo
← All terms

Related terms

Stable Diffusion

Stable Diffusion is a family of image generation models that create pictures from a text description. The first version was released in 2022, developed under the lead of Stability AI together with research partners. Technically it is a latent diffusion model that gradually turns image noise into a picture. Because the model weights are published, it can be run and customised locally; commercial use of newer versions is governed by Stability AI's own licence (as of October 2026).

  • image generation
  • diffusion model
  • +1

Negative prompt

A negative prompt specifies what an AI model should explicitly not generate or do. In image generation, for example with Stable Diffusion, it is a separate input field for unwanted elements such as “blurry” or “text in image”. With language models, it takes the form of rules in the instructions, such as “Do not make up prices” or “Do not give legal advice”. Such limits make results more predictable but do not replace review.

  • prompting
  • instructions
  • +1

Deterministic model

A deterministic model always produces the same result for the same input, because it follows fixed rules or formulas and involves no randomness – for example the rule '5% discount on orders over €100'. Large language models, by contrast, work probabilistically: their answers to the same question can vary. In practice the two are often combined to balance flexibility with traceability.

  • modelling
  • rules
  • +1

DALL-E

DALL-E is a series of image generation models from OpenAI that create images from text descriptions. The first version appeared in 2021; DALL-E 3 was integrated into ChatGPT in 2023 and made text-to-image generation accessible to a broad audience. OpenAI has since replaced DALL-E with newer image models from the GPT Image series; API access to DALL-E 2 and 3 ended in May 2026 (as of October 2026).

  • image generation
  • OpenAI
  • +1