贵模型编排便宜模型落地
Fable Plans Luna Implements
要点
- 用 fable-advisor 插件：Fable 5.1 做编排和复审，GPT-5.6 Luna 只执行。
- 提示词写明你是编排者，把实现委派给跑 luna 的 codex-implementer。
- 贵模型只做推理，便宜模型写代码，避免编排模型被实现细节拖垮。
原帖开头
This is my current workflow after Fable 5.1 dropped. Wired through the fable-advisor plugin for Claude Code: 1. Fable 5.1 orchestrates 2. GPT-5.6 Luna implements 3. Fable 5.1 reviews The prompt I run: "You are the orchestrator. Plan this task, then delegate implementation to …
