Zendesk
Freshdesk
Gorgias
Front
Shopify
Stripe
Slack
Notion
Google Workspace
Confluence
Large Language Model (LLM)
Definition
A large language model (LLM) is a neural network trained on vast amounts of text to predict and generate language, enabling it to understand questions, summarize, classify, and write human-like responses.
How it works
An LLM is built on the transformer architecture and trained to predict the next token (roughly, a word fragment) given the preceding text. By learning statistical patterns across billions of examples, it develops broad capabilities in comprehension, reasoning, and generation.
At inference time, you send the model a prompt and it generates a response token by token. Modern LLMs can also call tools and follow instructions, which is what makes them useful engines for AI agents.
Why it matters for support
LLMs are the reasoning engine behind conversational AI and AI agents. But out of the box an LLM only knows what it saw in training — it doesn't know your product or policies, which is why support systems ground it with retrieval-augmented generation and connected tools.
Frequently asked
What does LLM stand for?
Large language model — a model trained on large amounts of text to understand and generate language.
Do LLMs know my company's information?
Not by default. An LLM only knows its training data, so support systems feed it your knowledge base and live systems via RAG and tool calls to answer product-specific questions accurately.
Related terms
Generative AI
Generative AI is a class of AI that creates new content — text, images, code, or audio — by learning patterns from training data, rather than only classifying or predicting from existing inputs..
Foundation Model
A foundation model is a large AI model trained on broad, general-purpose data at scale that can be adapted to many downstream tasks — through prompting, retrieval, or fine-tuning — rather than being built for one narrow job..
Transformer Model
A transformer model is a neural network architecture that processes an entire sequence of text at once using an attention mechanism to weigh how much each word relates to every other word, and it's the foundation of nearly every modern large language model..
Tokens (LLM)
Tokens are the small chunks of text — words, parts of words, or characters — that a language model reads and generates; they're the fundamental unit models use to measure input, output, context limits, and usually billing..
Retrieval-Augmented Generation (RAG)
Retrieval-augmented generation (RAG) is a technique where an AI model retrieves relevant information from your own knowledge sources at query time and uses it to ground its answer, instead of relying only on what it memorized during training..
Put these ideas to work
Macha is an AI agent layer that sits on top of the help desk you already run — Zendesk, Freshdesk, Front, Intercom, or Gorgias.
Start Trial