Daniel & Friends / Glossary / Diffusion models
AI and automationDiffusion models
A diffusion model is a type of generative model that learns to remove noise from data. To create an image, it starts from random noise and builds the result step by step according to the prompt. Most image generators and many video generators rest on this principle.
Daniel Votruba ·
In practice
Settings such as the number of steps and prompt adherence affect quality and speed. For everyday marketing use, you just pick a finished tool and never handle the model itself.
Related terms
- Text-to-image: Text-to-image is the generation of images from a written prompt.
- Text-to-video: Text-to-video is the generation of short videos from a written prompt or from an image.
- Deep learning: Deep learning is a branch of machine learning that uses neural networks with many layers.
- Generative AI: Generative AI refers to models that create new content: text, images, audio, video or code.
More terms in the group AI and automation
Embedding · EU AI Act · Deepfake and synthetic media · Explainable AI (XAI) · Custom GPTs and projects · Few-shot and zero-shot prompting · Copilot (AI assistant) · Fine-tuning · Context window · Foundation model · Computer vision · Function calling (tool use) · ChatGPT Ads · Chain of thought · Grounding · C2PA and content provenance (AI watermarks) · Guardrails · Automation platforms (Zapier, Make, n8n) · Hallucination · AI workflow and orchestration · Human in the loop · AI slop · Inference and latency · AI search (answer engines)
Terms in practice
Need help putting them to work?
Terms are one thing, putting them to work is another. We go through which of them pay off for you and say so plainly. You get a first opinion for free and we reply within 24 hours.
Book a call