Glossary / Large language model
What is Large language model?
A large language model is a system trained on huge amounts of text that predicts likely next words, which is how it writes, summarises and answers.
A large language model, usually shortened to LLM, learns statistical patterns from an enormous body of text. Given some input it predicts what text plausibly comes next, one piece at a time.
That single mechanism explains both the capability and the failure mode. It is genuinely good at structure, tone, summarising and rephrasing, because those are patterns. It has no notion of truth, so a confident sentence about your product or a plausible statistic can be entirely invented.
For marketing work the practical rule follows directly: use it for things you can verify, and be the source of anything you cannot.
In Creaiter
Creaiter routes each request to the model that suits the job rather than sending everything to one, and you can bring your own provider keys.