What is a large language model (LLM)?
A large language model (LLM) is an AI trained on vast amounts of text to understand and generate human-like language, powering chatbots, assistants, and AI agents.
LLMs learn patterns from enormous text datasets, letting them answer questions, write, summarize, translate, and reason. Examples include the models behind ChatGPT, Claude, and Gemini.
On their own, LLMs generate text. To take actions in the world, they're paired with tools, memory, and an agent loop - which is how a model becomes a useful assistant rather than just a smart autocomplete.
how this works in folk
Folk uses leading LLMs (from Anthropic, OpenAI, and others) as its reasoning engine, then adds a cloud computer, memory, and integrations on top - and you can bring your own model and keys.
frequently asked
What's the difference between an LLM and an AI assistant?
An LLM is the underlying model that generates language. An AI assistant wraps a model with memory, tools, and an interface to actually help you. Folk is the assistant; the LLM is one part of it.
Which LLM does folk use?
Folk uses the best available models and lets you bring your own Anthropic, OpenAI, or OpenRouter keys.