LLM (Large Language Model)
An LLM (Large Language Model) is an artificial intelligence model trained on vast text corpora to understand and generate natural language. ChatGPT, Claude, Gemini and Llama are examples: they power the answer engines reshaping search.
An LLM (Large Language Model) is an artificial intelligence model trained on billions of words to predict and generate natural language. It's the technology behind ChatGPT, Claude, Gemini and Llama. An LLM doesn't "search" like a classic engine: it composes an answer from what it has learned and, increasingly, from sources consulted live.
How an LLM works
An LLM learns the statistical regularities of language during training, up to a cutoff date. To answer a question, it generates the most probable text given what it has learned. When connected to web search or a RAG system, it supplements its memory with recent documents — and that's what makes citing your content possible.
A concrete example
When you ask Perplexity "what is the best GEO agency in France?", the underlying LLM assembles an answer from pages it consults in real time and from what it has memorized about these brands. Your presence in both counts.
Why it matters
Understanding how an LLM works means understanding why LLMO and GEO rely so heavily on clear definitions and brand mentions. To put this lever to work, see our GEO agency.
Cutoff date
The end date of an LLM's training data. Beyond it, the model depends on web search to stay up to date.
Questions fréquentes
Not by default: an LLM stops at its training cutoff date. To answer on recent topics, it relies on web search and RAG, which fetch up-to-date sources at the moment of the question.
Termes & ressources liés
Une question sur votre visibilité IA ?
Score de visibilité IA de votre site. Gap analysis vs 3 concurrents directs. 5 optimisations prioritaires. Livré en PDF, sans engagement.
Réponse sous 24h · Sans engagement · contact@luwiz.io