A practical token budget rulebook for LLM agents
In LLM , the cost question is not only which model to use. The bigger issue is how to spend tokens and expensive carefully. The main goal is to save the strongest budget for decisions that truly need it.
Broad research and mechanical work should go to the cheapest capable model, often in parallel, with only narrow and structured results passed upward. Final judgment should come together in one place instead of being scattered across many smaller calls. The workflow should summarize once and reason once, rather than repeatedly compressing and rethinking the same material.
should happen step by step, so the system brings in only the information it needs instead of loading at the start. When something is unclear, the system should raise the to a stronger decision step instead of letting a cheaper model guess. These are general principles, not a fixed ranking of models.
Key points
- Keep expensive for decisions that truly need it.
- Use the cheapest capable model for routine tasks like sorting, formatting, and gathering evidence.
- Run broad work in parallel, then pass only structured results to the main decision step.
- Use gradually instead of loading all source material at once.
- Escalate unclear cases rather than letting a cheaper model guess.