Models & ProductsLLM Architecture Updated 2026

Diffusion Model

Definition

A generative model that learns to create data by reversing a gradual noising process — the technology behind most modern AI image generation.

Diffusion models, framed by Sohl-Dickstein et al. (2015) and made practical for high-quality image synthesis by Ho et al. (2020), work in two phases: a forward process that progressively adds noise to data, and a learned reverse process that denoises random noise back into a coherent sample.

They underpin today's leading text-to-image systems and are increasingly used for audio and video generation.

References

Primary, peer-reviewed and archival sources for this definition.

Denoising Diffusion Probabilistic Models
Ho, J., Jain, A., & Abbeel, P. (2020). Advances in Neural Information Processing Systems 33 (NeurIPS 2020).
Deep Unsupervised Learning using Nonequilibrium Thermodynamics
Sohl-Dickstein, J., Weiss, E. A., Maheswaranathan, N., & Ganguli, S. (2015). Proceedings of the 32nd International Conference on Machine Learning (ICML 2015).

Dictionary & encyclopedic entries

Cite this entry

MultipleChat. "Diffusion Model." MultipleChat AI & LLM Glossary, 2026. https://multiple.chat/ai-glossary/diffusion

Related terms

See this in practice

Run the same prompt across ChatGPT, Claude, Gemini, Grok and Perplexity side by side — with your own files as context and answers you can cross-check.

Try MultipleChat Free

Continue learning

See paid plans