Systems session 1/3 · Exercises

Tokenizers and token budgets

  1. Estimate a 4,000-character prompt with the approximation tool.
  2. Measure it with the intended model tokenizer.
  3. Explain the difference and reserve output headroom.
Show answer

Yes. Vocabulary, normalization, and segmentation rules differ by tokenizer. The complete serialized request plus the output tokens reserved for generation.