This workflow compares two editing strategies in Nano Banana 2.1—prompt-only global edits versus mask-guided local edits—using a single model node (GeminiNanoBanana2V2) and a consistent pipeline. It loads your image (e.g., the provided silver_frame_profile.png), applies your natural-language edit, and renders a side-by-side before/after via ImageCompare so you can judge impact at a glance. Running once without a mask showcases broad, scene-wide transformations; running with a mask constrains changes to selected regions while preserving the rest.
Methodologically, you keep the same prompt and parameters and only toggle whether a mask is supplied to GeminiNanoBanana2V2. This isolates the effect of masking: global edits offer cohesive stylistic shifts but risk drifting subjects and text, while masked edits improve control and identity retention but may show edge artifacts if the mask is hard. SaveImageAdvanced preserves versioned outputs, and MarkdownNote helps you log prompts, mask intent, and observations across runs for repeatable, defensible comparisons.
API
從程式碼使用此工作流
每個 Comfy 工作流都是一個 JSON 圖。下方的 payload 就是此工作流，與 ComfyUI 執行時完全相同 — 你可以從此 URL 取得、放入版本控制，或在 ComfyUI 載入並逐節點執行。
使用 Comfy SDK 以 TypeScript 或 Python 執行。相同程式碼可用於 Comfy Cloud 或你自架的 ComfyUI — 只需更改 base URL。
// Install (beta)
npm i @comfyorg/sdk
// Run this workflow (TypeScript)
import { Comfy } from "@comfyorg/sdk";
const client = new Comfy({ apiKey: "comfyui-..." });
const wf = await client.workflows.fromFile("workflow_api.json");
const job = await client.run(wf);
await job.getOutputs("<output-node-id>")[0].toFile("output.png");SDK 需使用 API 格式的工作流：請在 ComfyUI 開啟此工作流，並使用 檔案 → 匯出工作流（API）。
