Glossary
Large language model
What a large language model is, and what it needs from a business before it can answer that business's customers.
Updated
Large language model: A large language model, or LLM, is the kind of AI that reads and writes plain language. It learns from a vast amount of text until it can predict the words most likely to come next in whatever it is given, which is how it answers a question or drafts a reply. It knows only what that text held, so about one small business it may know nothing, or something out of date. To answer for a business, it has to be given the business's own pages and records as it works, and held to them.
Where it shows up
You meet the term on the pages of AI products, which often name the model they run on, and on the bills of the companies that make the models. They usually charge by the amount of text a model reads and writes, counted in tokens, which are small pieces of words. The same model can sit behind very different products, because what each one does depends on what it is connected to and the rules it follows.
Each AI job we build runs its model on an account in your name and has it answer from the sources your written plan names, such as your site's pages or one system's records, with the usage billed to you at cost on that account.
Related terms
- AI hallucinationWhat an AI hallucination is, and how an AI job is set up to catch one.
- APIWhat an API is, and why it decides which of a business's tools an AI job can connect to.
- Answer engineWhat an answer engine is and how it changes the way customers find a small business online.
- Every term in the glossary
Start with a written plan
Tell us about the business and what you need. We reply with questions, then a written plan that lists the work and its price, and nothing starts until you have read it.