Do I need local Stable Diffusion models for this workflow?

No. The core generation is handled by GeminiImage2Node via an external Google image API. You just need a valid API configuration; no local checkpoints are required.

How do I keep features perfectly consistent across views?

Use the same reference image every time and keep prompts tightly scoped: specify “orthographic,” “no perspective,” and call out key elements to match (helmet shape, plating, decals, color). If results drift, tighten language (e.g., “exact same chest plate geometry and markings as the reference”) and keep background simple.

Can this make a single sheet with all views automatically?

This template saves each view as a separate file via SaveImageAdvanced. You can assemble them in your editor, or add a tiling/composition step in ComfyUI if you prefer an all‑in‑one sheet.

Will it work for organic characters or only mechs?

It works for both, but hard‑surface designs tend to hold consistency more easily. For organic characters, be extra explicit about proportions, clothing details, and pose (neutral T‑pose, arms down, etc.).

Character Turnaround Sheet Generator

This workflow generates a clean, orthographic character turnaround (front, side, and back) from a single reference image—ideal for sci‑fi mechs and armor designs. It starts with LoadImage to bring in a front orthographic reference (a sample white_mech_front_ortho.png is included). That image is then passed to GeminiImage2Node, which calls a Google Generative image API and uses your text instructions plus the reference to synthesize additional angles while preserving key features like silhouette, plate layout, decals, and color scheme. SaveImageAdvanced writes each view to disk with predictable filenames so you can assemble them into a final sheet in your design tool. A MarkdownNote in the graph provides inline guidance and prompt tips.

Technically, consistency across views is encouraged by conditioning GeminiImage2Node with the same reference image and carefully worded prompts (e.g., “orthographic side view of the same mech as the reference; match plating, helmet, and markings; no perspective; white background”). Because this template relies on an external API rather than local checkpoints, it’s quick to set up—no model downloads are required—making it a practical tool for concept artists and product designers who need fast, multi‑view explorations of a single design.

API

Use this workflow from code

Every Comfy workflow is a JSON graph. The payload below is this workflow, exactly as ComfyUI runs it — fetch it from the URL, keep it in version control, or load it in ComfyUI and run it node by node.

ad315e41d8c2.json

Fetching workflow JSON…

Run it from TypeScript or Python with the Comfy SDK. The same code targets Comfy Cloud or a ComfyUI you host yourself — only the base URL changes.

// 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");

The SDK takes a workflow in API format: open this workflow in ComfyUI and use File → Export Workflow (API).

Comfy Cloud API access requires a plan with an API key.

SDK docs Get an API key

FAQ

Frequently Asked Questions