Load the corpus·2 minTokens vs characters0 / 1Knowledge checkWhy does this course budget context in tokens rather than characters?ATokens are easier to countBThe LLM's context window and cost are measured in tokens, and tokens-per-char varies by text, so a character budget would over- or under-fill the real windowCCharacters can't represent unicodeDEmbeddings require token counts Load the corpusWhy chunk at all