Daniel & Friends / Glossary / Embedding
AI and automationEmbedding
An embedding is a numerical representation of text, an image or other data as a vector, where similar meaning means a nearby position. That lets a system see that “returning goods” and “complaint about a purchase” are related, even without a shared word.
Daniel Votruba ·
In practice
Embeddings are created when you prepare data for search and RAG. Stick with one embedding model across the whole set, because vectors from different models cannot be mixed.
Related terms
- Vector database: A vector database stores embeddings and quickly finds those closest to a given query.
- Semantic search: Semantic search finds results by the meaning of a query instead of matching individual words.
- RAG (retrieval-augmented generation): RAG is a method in which a model first retrieves relevant documents, for example company knowledge, and answers from them.
- Large language model (LLM): A large language model is a model trained on a huge amount of text that predicts the next words and can write, summarise and answer.
More terms in the group AI and automation
EU AI Act · Diffusion models · Explainable AI (XAI) · Deep learning · Few-shot and zero-shot prompting · Deepfake and synthetic media · Fine-tuning · Custom GPTs and projects · Foundation model · Copilot (AI assistant) · Function calling (tool use) · Context window · Generative AI · Computer vision · Grounding · ChatGPT Ads · Guardrails · Chain of thought · Hallucination · C2PA and content provenance (AI watermarks) · Human in the loop · Automation platforms (Zapier, Make, n8n) · Inference and latency · AI workflow and orchestration
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