LLM (large language model)
An LLM is the AI model that understands and writes text, such as the models behind ChatGPT, Claude and Gemini. It is the engine under an AI agent, not the agent itself.
A language model is trained on enormous amounts of text and uses that to predict the most likely continuation. That lets it summarise, translate, classify and draft. It knows nothing about your company unless you give it that information, through the prompt, a knowledge base (RAG) or a connection to your systems.
For a company the model choice is less exciting than it seems. What matters more is what surrounds it: which data the model sees, where it runs (EU or not) and who checks what comes out.
Other terms
RAG (retrieval-augmented generation)
RAG is the technique where an AI model first looks up the relevant passages in your own documents and answers with those, so the answer is based on your information and has a source.
Vector database
A vector database stores text as series of numbers that capture its meaning, so you can search for "what resembles this" instead of only exact words.
MCP (Model Context Protocol)
MCP is an open standard through which AI models and agents access tools and data sources in a uniform way, so an integration is built once and usable by any agent.
Want to know what this means for your business?
Tell us where you lose time or margin. We show where AI pays back fastest, in plain language.