Problem

Complex tasks require careful planning but straightforward execution. Using an expensive model for all steps is wasteful; using a cheap model to plan produces poor plans.

Solution

Use a powerful model (GPT-6.1 Sol, Claude Sonnet 5.5) to generate a complete step-by-step plan, then execute each step with a cheaper, faster model (GPT-6 Luna, Claude Haiku 5.5).

Implementation

  1. Plan: Call the planner model with: "Generate a detailed step-by-step plan to [goal]. Return as a JSON array of steps."
  2. Store plan: Persist the plan; allow human review if needed
  3. Execute: For each step, call the executor model with the step description + context from previous steps
  4. Replan trigger: If a step fails or produces unexpected results, call the planner model to revise the remaining plan

Trade-offs

  • Pro: Lower cost than running every step on the expensive model; the saving depends on how much of the work the executor handles
  • Pro: Plan is inspectable and modifiable before execution
  • Con: Planner-executor mismatch: planner may assume capabilities the executor lacks
  • Con: Replanning adds latency

When to use

Research tasks, code generation with multiple files, data analysis pipelines, content creation workflows.

When to avoid

Simple tasks where a single prompt suffices; tasks with highly dynamic, unpredictable steps.