Özer Cem KELAHMETKELAHMET
Glossary · Deep learning

What is Diffusion models? Definition and meaning

Generative models that learn to denoise data step by step, turning noise into a sample.

Definition

Diffusion models: Generative models that learn to denoise data step by step, turning noise into a sample.

In short: Generation by learning to reverse a noising process.

When it is used

Image, audio and video generation, and increasingly structured data synthesis.

Where it fits

Diffusion models belongs to Architectures in the Deep learning topic. These are the other terms in the same group:

  • CNN: Networks that apply learned filters across local neighbourhoods, sharing weights over positions.
  • Recurrent neural networks: A neural network that processes a sequence one step at a time, updating a hidden state that summarises everything seen so far.
  • LSTM: A recurrent network with gates that control what is stored, forgotten and emitted over time.
  • GRU: A recurrent unit with fewer gates than an LSTM, often similarly accurate and faster to train.
  • Transformers: Neural networks built on self-attention, letting every element of the input weigh every other element.
  • ResNet: A convolutional architecture whose residual connections make very deep networks trainable.
  • ViT: Splits an image into patches and processes them as a sequence with a transformer.
  • Autoencoders: A network that encodes an input into a smaller code and decodes it back, learning what matters.
  • VAE: An autoencoder that learns a distribution over a latent space and is trained with a variational objective.
  • GAN: Two networks compete: a generator produces samples and a discriminator tries to tell them from real data.

Learn more

The full guide to Diffusion models explains how it works, when to use it and how it compares with the alternatives, with examples. The glossary lists every term on the site from A to Z.

Previous term: Diffusion language models · Next term: Digital twins

Last updated Nov 18, 2026
Type to search.
    ↑↓ navigate↵ open