What kind of reference image works best?

Use a high‑quality image with a clear subject, strong silhouette, and minimal background clutter. Center the subject, avoid heavy motion blur, and keep lighting consistent—these factors help the node preserve appearance across frames.

How do I enhance the stop‑motion look?

Lower the FPS in SaveVideo (around 8–12), keep motion amount modest, and avoid excessive smoothing. Shorter clips with small, discrete movements feel more like physical stop‑motion than long, fluid camera moves.

How do I control video length and size?

Set clip length via the duration/frames controls in ByteDance2ReferenceNode and set playback speed with FPS in SaveVideo. Use an image resolution that fits your GPU memory; larger frames cost more VRAM and time.

Why does the subject drift or deform between frames?

This often happens when motion is set too high or the image has weak structure. Reduce motion amount, use a cleaner/higher‑res reference, and increase how strongly the node adheres to the reference if that control is available.

Seedance 2.0 Reference to Video - Concept Art + Stop Motion Style

This ComfyUI tutorial workflow turns a single piece of concept art into a short, stop‑motion‑style video. It uses LoadImage to bring your reference artwork into the graph, feeds it into ByteDance2ReferenceNode to synthesize motion while preserving the look of the original design, and then encodes the result with SaveVideo. The ByteDance2ReferenceNode is the engine of the pipeline: it performs reference‑to‑video generation by conditioning on your image so the character, style, and palette remain consistent across frames while introducing controlled movement.

Technically, the workflow keeps things simple and focused on iteration. You supply a clean, well‑framed reference image; ByteDance2ReferenceNode generates a sequence of frames guided by the reference, and SaveVideo writes those frames to an MP4 at a frame rate you choose. To emphasize a stop‑motion aesthetic, you typically use a lower FPS (for example, 8–12) and modest motion magnitude so changes appear stepwise and tactile. Because the setup is lightweight (LoadImage → ByteDance2ReferenceNode → SaveVideo), it’s easy to re‑run with different reference images or parameter tweaks to dial in timing, motion intensity, and framing.

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.

70e6b211c5ff.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