How long can the generated video be?

Length is determined by the number of frames and FPS. For example, 48 frames at 12 FPS ≈ 4 seconds. Longer clips require more VRAM and time; increase frames gradually and consider lowering resolution or FPS if you hit memory limits.

How do I keep the video faithful to the original image?

Use a conservative prompt and reduce the node’s motion/denoise strength so it deviates less from the source. Also try increasing guidance strength and using a fixed seed. If available, choose a sampler or setting that favors stronger conditioning.

My output has jitter or artifacts. What should I change?

Lower motion/denoise strength, reduce resolution, and try a different seed. You can also decrease the number of frames or use a slightly higher guidance value. Subtle, specific prompts often produce more stable motion than broad or conflicting ones.

Which formats can I save to?

SaveVideo can assemble the frames into a video container like MP4 or WebM (options depend on your node’s settings and installed encoders). Match FPS between the generator and SaveVideo to avoid timing issues.

Image to Video - ComfyUI Workflow

This ComfyUI workflow turns a single still image into a short video guided by a text prompt. It uses LoadImage to bring in your source frame, passes it to a dedicated Image-to-Video generator node (ID: 770dde0e-f342-49aa-ad1e-184724671c5d), and then encodes the resulting frame sequence with SaveVideo. The generator node conditions on both the input image (as a visual anchor) and your prompt (to suggest motion, style, or scene changes), producing a coherent set of frames that evolve over time.

Technically, the generator node synthesizes a batch of images (frames) from the static input by sampling with a controllable seed and typical parameters such as number of frames, motion/denoise strength, guidance, and resolution. SaveVideo then assembles those frames into a playable clip at your chosen FPS and output format (e.g., MP4 or WebM). You control duration via frame count and playback feel via FPS, while the seed lets you repeat or iterate on results consistently. This makes the workflow useful for quick motion studies, cinemagraph-like clips, and social-ready animations without needing full video capture.

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.

3515c5083027.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

View all workflows