Tokens
The chunks of text models bill and process. Context windows are in tokens. Your page might get truncated in RAG.
Put the answer early. A 20k-token waffle loses the ending.
Token cost is why engines cache and skip retrieval.
Examples
- The important caveat is at the bottom of a huge HTML page. The chunker never reached it.
Related terms
Inference is running the trained model to produce an answer. Every ChatGPT reply is inference. It costs money, which is why vendors cache and skip retrieval.
Chunking is writing sections that still make sense if a retriever lifts 100–300 words and ignores the rest of the page.
How much text a model can consider at once. Bigger windows mean longer threads and more pasted docs — including your pricing PDF.
FAQ
Shorter pages always better? +
Shorter sections. The site can still be long. Don’t hide the lead.
Track this in Reddex
See Reddit threads and AI answers for your brand in one place. Start with a free analysis.
Get started