AI glossary · Models
LLM
In one sentence
An LLM (large language model) is an AI model trained on vast amounts of text to predict the next word. That single skill lets it draft, summarise, translate, classify and hold a conversation.
Also known as: large language model · foundation model · language model
What is an LLM?
LLM stands for large language model. It’s a language model set apart by its scale: billions of parameters tuned during training on books, web pages, code and other text. That scale is what makes it fluent on almost any topic.
Put it to work
AI agents that look after your customers
From €49/mo + VAT
Voice, WhatsApp and email with the phone line included: agent, numbers and CRM in one fee.
See plans and pricingAn LLM doesn’t look facts up in a database. It learns patterns: which words tend to follow which, how an email is structured, how people respond to a complaint. That’s why it writes so well, and also why it can be confidently wrong. The term for that failure is hallucination.
Well-known examples include the GPT, Claude, Gemini, Llama and Mistral families. Versions change often, so understanding how they work matters more than keeping track of names.
Related to this term
UTIO agents
- Available 24/7 in 26 languages
- Set it up yourself for €0: built from your website
- 14-day free trial, no card needed
From €49/mo + VAT
See plans and pricingHow it works
- Split the text into tokens, a step called tokenization.
- Process those tokens with a transformer network, which works out which parts of the text matter for each word.
- Predict the most likely next token, append it and repeat until the reply is complete.
How much randomness goes into each choice is set by the temperature. Everything the model considers in one request has to fit in its context window: your instructions, any documents you pass in and the conversation so far.
Example
An accountancy practice in Leeds gets dozens of emails a day asking similar things: VAT deadlines, paperwork for registering as self-employed, moving an appointment. An LLM can read each email, work out what’s being asked and draft a reply in seconds.
What the LLM doesn’t know is the practice’s own information: opening hours, fees, where each client’s file stands. So it’s paired with RAG, which hands it the exact passage from the practice’s documentation for each query. A member of staff checks the draft before it goes out.
LLMs at UTIO
UTIO’s agents use an LLM to understand what your customer says or types and to decide how to respond. With the voice agent and the WhatsApp agent, you don’t pick a model or count tokens: it’s part of the monthly fee. The language model runs on our own servers, with fallback to European and US platforms under contractual safeguards.
If you’re building your own application, the AI API offers text generation at the model tier you choose: basic, standard or superior. That way the LLM becomes the core of an AI agent rather than an end in itself.
Frequently asked questions
Is ChatGPT an LLM?
Not quite. ChatGPT is a chat application that runs on LLMs from the GPT family. The LLM is the engine that produces the text; the app adds the interface, chat history, tools and usage rules.
Does an LLM search the web when I ask it something?
Not on its own. It answers from what it learned in training, which stops at a cut-off date. If an application gives it a search engine or your documents, it can look up current information, but that comes from the system built around it.
What's the difference between an LLM and an AI agent?
An LLM generates text. An AI agent uses an LLM as its brain and adds a goal, tools and rules so it can get things done, such as booking an appointment or updating a CRM.
Related terms
- Language model
A language model is an AI system that calculates which word, or token, is most likely to come next in a piece of text. By repeating that prediction it can complete sentences, answer questions, summarise or translate.
- Transformer
A transformer is a neural network architecture that processes every word in a text at once and uses an attention mechanism to work out which parts relate to each other. It's the foundation of almost every language model in use today.
- Token
A token is the chunk of text a language model works with: a word, part of a word or a punctuation mark. Models read, count and bill in tokens, not in words or characters.
- AI hallucination
An AI hallucination is a response from an AI model that sounds convincing but is false or not backed by any source: a figure, quote, price or reference the model invents while generating the most probable text.
- Small language model (SLM)
A small language model (SLM) is a language model with far fewer parameters than an LLM. It knows less about everything, but it replies faster, costs less and can run on a modest server, a laptop or even a phone.
Updated on