# Large language model (LLM)

A large language model, or LLM, is an AI system trained on enormous amounts of text data to understand and generate human language, predicting what a coherent, contextually appropriate response looks like.

It's the technology behind most modern chatbots, writing assistants, and the language understanding inside AI interview systems.

**Large language model (LLM)** — A large language model, or LLM, is an AI system trained on vast amounts of text to predict and generate coherent, contextually relevant language. During training, the model learns patterns across huge volumes of text, which lets it generate a reasonable next response given almost any input, from answering a question to summarizing a document to evaluating whether a spoken answer actually addressed what was asked. Unlike older, narrower NLP tools built for one specific task, a single LLM can handle a wide range of language tasks, which is why LLMs have become the underlying engine behind most current conversational AI, automated writing tools, and language-based scoring systems. Different LLMs vary in size, training data, and cost to run, and a given product typically picks a model suited to its specific task rather than always using the largest one available.

In hiring, LLMs sit behind a range of tools: generating interview questions, understanding and following up on a candidate's spoken answer, summarizing a long interview transcript, and producing written reasoning behind a numeric score. The same underlying model type can do all of these because the common task, understanding and generating language, doesn't fundamentally change across them, even though the specific prompt and output format does.

## Frequently asked questions

### Is an LLM the same thing as ChatGPT?

ChatGPT is one product built on top of an LLM. The term LLM refers to the underlying model type, of which several exist from different companies, rather than any single named product.

### Do LLMs understand language the way a person does?

Not in the human sense. An LLM predicts likely, contextually appropriate language based on patterns learned from its training data, which can produce very humanlike output without the model having genuine understanding the way a person does.

### How does an LLM get used in AI interview scoring?

An LLM can read an interview transcript, evaluate how well an answer addressed the question, and generate both a score and written reasoning explaining that score, based on patterns it learned about what strong and weak answers look like.

## Related pages

- [Natural language processing (NLP)](/glossary/natural-language-processing)
- [AI interview scoring](/glossary/ai-interview-scoring)
- [Browse open roles](/jobs)

## See language AI applied to a real interview

Intervieux uses language models to grade real interview answers with written reasoning behind every score, not a fixed keyword checklist.

Start practicing free: https://www.intervieux.ai/register · Hire with Intervieux: https://www.intervieux.ai/employers/signup