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
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.”
Created: 8/13/2026, 9:50:58 AM
My Notes
Loading notes...