Large Language Models (LLMs)
Neural networks trained on massive text corpora that generate human-like text.
Advanced AI systems trained on vast amounts of text data that can understand, generate, and manipulate human language with remarkable fluency and versatility.
LLMs are the engine behind ChatGPT, Claude, Gemini, Perplexity, and most modern AI tools. Each model has slightly different appetites for source material; sites that publish multiple structured formats cover the union of those appetites.
Large language models are the engines behind ChatGPT, Claude, Gemini, Perplexity, and the AI features now embedded in search. Because these systems read, summarize, and answer from web content, being legible to them has become its own discipline: content that is clearly written and well-structured is easier for a model to parse, ground, and cite accurately. As more discovery happens through LLM-mediated answers, optimizing for how these models consume content complements traditional search optimization.
- Answer generationLLMs synthesize responses from multiple sources, making inclusion as a cited source the new visibility goal.
- Summarization and extractionModels condense pages into short answers, rewarding content that states key facts clearly and early.
- Conversational assistantsAgents built on LLMs read the web on a user's behalf, favoring content they can parse and reuse reliably.
LLMs process content as sequences of tokens, weighting the parts that look most relevant to a query. They handle clearly-structured, fact-dense text more reliably than long, ambiguous prose, and - because each model has a finite context window - compact, well-organized content is easier to include and quote. Publishing content in clean, machine-readable form helps ensure the version a model sees matches what you intend.
Knowing the term is step one.
Geordy operationalizes every term in this glossary - generating the structured files AI engines actually read.