Deploy a photo-to-video ComfyUI workflow on a cloud GPU
Run a production-ready Wan 2.2 image-to-video workflow without rebuilding the Docker image, hunting custom nodes, or hand-scripting model downloads.
Start from a specific workflow outcome, review the expected GPU and cost profile, then deploy the hosted ComfyUI template when the fit is right.
- Workflow
- Wan 2.2 5B I2V
- GPU
- A6000 48GB
- First deploy
- ~15 min
- Per video
- ~$0.10
Why this deployment path works
The page is built around a specific workflow outcome, a real template, a visible demo artifact, and a direct path into the hosted deployment flow.
Starts from a tested template
Photo Animator is already in the production-ready template set and points to the existing deployment flow.
Keeps the workflow editable
The deployment opens ComfyUI, so the user can adjust prompts, inputs, seeds, and workflow settings after launch.
Makes cost visible before signup
The page states the expected GPU tier, first-deploy window, and rough per-video cost before pushing the deploy CTA.
End-to-end path
Move from a clear creative intent to a hosted workflow without rebuilding the ComfyUI environment from scratch.
- 1
Open the template
Start from the Photo Animator template instead of a blank GPU pod.
- 2
Deploy the GPU workspace
ModelPilot provisions the A6000 deployment and prepares the workflow environment.
- 3
Upload a photo in ComfyUI
Use the hosted ComfyUI interface to add the input image and prompt.
- 4
Generate, review, then stop
Create the clip, save the output, and stop the deployment when finished to control spend.
Use this workflow when
- You have product photos, portraits, or portfolio images to animate
- You want the ComfyUI graph available after deployment, not a black-box API
- You want a reusable GPU workspace instead of rebuilding setup from scratch
Choose another path when
- You only need a few low-stakes images and do not care about ComfyUI
- You need fully automatic bulk video generation with no manual review
- You require a custom model license or private LoRA audit before launch
Frequently asked questions
Is this a generic image-to-video API?
No. This is a hosted ComfyUI workflow deployment. You get a GPU-backed ComfyUI workspace with the Photo Animator workflow rather than a fixed black-box endpoint.
Can I use my own workflow instead?
Yes. Use the workflow compatibility checker first. It analyzes custom nodes, model references, and likely GPU requirements before deployment.
Why not set this up directly on RunPod?
You can. The tradeoff is time: DIY setup requires selecting a template, resolving custom nodes, downloading model files, and wiring ComfyUI yourself. ModelPilot charges a service layer for that managed path.
Ready to deploy the workflow?
Use the hosted Photo Animator template when you want an editable ComfyUI workspace instead of another setup guide.