Do I need two images, or can I use just one?

You can use one or two. One image will guide both the start and end, creating motion that returns to the original framing. Adding a second image as the last frame lets Wan2.7 transition toward a different final composition.

How does the audio affect the output?

The audio defines the video’s duration and helps align motion pacing. Choose a track with the length you want. If the node exposes timing controls, keep them consistent with the audio to avoid truncation or extra silence.

What resolution and aspect ratio should I use?

Match the aspect ratio between your first and last frames, and keep a reasonable resolution your hardware or API quota can handle. Higher resolutions need more compute and time—start smaller to validate motion, then scale up.

How can I get cleaner, more stable motion?

Use sharp, well-lit images with consistent subject size and framing across the first and last frames. Avoid drastic changes in viewpoint or lighting between endpoints. If available, reduce motion intensity settings for subtler moves.

Wan2.7: Image to Video - ComfyUI Workflow

This ComfyUI tutorial workflow shows you how to turn a still image (or a pair of start/end images) and an audio track into a coherent video with sound using Wan2.7. The core is the Wan2ImageToVideoApi node, which takes your reference image(s) and audio, then synthesizes the in-between frames so the video begins on your first frame, transitions naturally, and lands on your last frame. LoadImage nodes feed the first and last control frames, while SaveVideo writes the final output to disk.

Technically, the first and last frames act as strong compositional anchors: Wan2.7 preserves the subject and layout at those endpoints and generates motion between them. The audio you provide is used to align timing and pacing so the video duration matches the track and motion evolves in sync. Keep your start and end images aligned in aspect ratio and subject framing for cleaner transitions. The SaveVideo node then packages the generated frames (and the synchronized audio from the API) into a playable video file.

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.

api_wan2_7_i2v.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 "Wan2.7: Image to Video" (TypeScript)
import { Comfy } from "@comfyorg/sdk";

const client = new Comfy({ apiKey: "comfyui-..." });

// This workflow, exported in API format (see note below)
const wf = await client.workflows.fromFile("api_wan2_7_i2v_api.json");

const asset = client.assets.fromFile("input.png");
wf.setInput("24", "image", asset);  // LoadImage

const job = await client.run(wf);
await job.getOutputs("32")[0].toFile("output.png");  // SaveVideo

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