JSON and other structured formats consume more tokens than plain text because they include repeated declarations of labels and schemas that serve as machine-parsing guidance but represent wasted token consumption for the LLM.

factualpending

Speaker

Unidentified Speaker — A Quick Lesson Of AI Languages [y1NBC1iXiL4]

Evidence Quote

JSON has repeated declarations of certain labels or certain schemas... for the LLM, that's just another token. That's another token that fills up this window.

Source

A Quick Lesson Of AI LanguagesShane Murphy
Created: 8/13/2026, 9:50:58 AM

My Notes

Loading notes...