WAN 2.2 Replace AI Generator
Generated on PixelDojo with WAN 2.2 Replace. Produced by PixelDojo's generation pipeline.
WAN 2.2 Replace costs 2 credits per second of output, so a 5 second swap runs 10 credits, and ours finished in about 110 seconds on August 26, 2026. It takes a source video and a single character image, then drops that character into the existing scene while keeping the background and camera motion intact. This page covers the price, what our swap actually looked like, and how Replace differs from its sibling, Animate.
The barista clip with a swapped-in character
Every example below was produced on PixelDojo. Hover to see the prompt.
Replace the character in this video with the reference character
WAN 2.2 Replace
What the run showed
2 credits per second, rounded up
video_duration is meant to run 1 to 15 seconds and should not exceed the source clip's own length. We requested 5 seconds and were charged 10 credits, matching the formula.
It needs two inputs, not a prompt
video (the source clip) and character_image (the replacement face and body) are both required. There is no text prompt field on this endpoint at all.
The output ran longer than the requested duration
We asked for a 5 second result but the file we downloaded measured about 8 seconds, matching the length of our 8 second source clip rather than the video_duration we set. Treat that field as a ceiling the model does not always enforce, not a guarantee.
The swap preserved the scene convincingly
We fed it a cafe clip of a barista pouring latte art and a portrait of a weathered fisherman as the replacement character. The result kept the espresso machine, the golden backlight, and the pouring motion exactly where they were, with the poured character now showing the fisherman's older, more weathered hands and face.
Source audio carries through when you ask for it
merge_audio defaults to true on this endpoint, and our output kept a real audio track rather than coming back silent.
Why Choose Pixel Dojo for WAN 2.2 Replace
Professional-quality results with cutting-edge AI technology
2 credits per second, straightforward math
No resolution surcharge baked into the base rate at 480p. Cost is just the seconds you request, rounded up, times 2.
Keeps the original scene and motion
Background, lighting, camera movement, and the action itself all carried over from the source clip in our test. Only the character changed.
How It Works
How we ran it, so you can repeat it:
Provide a source video and a character image
We used an 8 second cafe clip as the video and a portrait as the character_image. No prompt field exists on this endpoint.
Set video_duration and resolution
We requested 5 seconds at 480p, the cheapest resolution tier, for 10 credits.
Check the returned length against your source, not just your setting
Ours came back closer to the 8 second source than the 5 seconds requested, in about 110 seconds.
Loved by creators on PixelDojo
Real feedback from people using PixelDojo, pulled from our in-product surveys.
Awesome site with so many features
Love how I can almost create anything
The Flux Pro Ultra is just amazing!
Very easy to use, and they have fast wan 2.2 video generation with custom loras available.
The overall quality of the site and it's amazing variety of tools. The continued updates of the UI. The unbelievable level of tech support.
THIS IS SO DOPE !
Explore more AI tools on PixelDojo
AI Tools
Compare & Switch
- Best AI Image Generators
- Best AI Video Generators
- Midjourney Alternatives
- Civitai Alternatives
- Runway Alternatives
- Leonardo Alternatives
- Pika Alternatives
- Luma Alternatives
- Magnific Alternatives
- Veo Alternatives
- Flux Alternatives
- Freepik Alternatives
- Seedance Alternatives
- Seedream Alternatives
- Pixverse Alternatives
- GPT Image Alternatives
- Synthesia Alternatives
- Playground Alternatives
- NightCafe Alternatives
- Canva AI Alternatives
- ElevenLabs Alternatives
- ComfyUI Alternatives
- Fal Alternatives
- Replicate Alternatives
Common Questions
Everything you need to know about WAN 2.2 Replace
What does WAN 2.2 Replace cost?
2 credits per second of the video_duration you request, rounded up. A 5 second swap is 10 credits.
Does the source video length affect the price?
No, only video_duration does, though the tool's own guidance says that value should not exceed your source clip's real length. In our test the returned file ran closer to the 8 second source than the 5 seconds we requested, so budget for the source length, not just the field you set.
How fast is WAN 2.2 Replace?
About 110 seconds in our August 26, 2026 run for a 5 second request against an 8 second source clip. Real turnaround moves with queue load.
How do I call it from the API?
POST to /api/v1/models/wan-2.2-replace/run with your API key, a video URL for the source clip, a character_image URL, and video_duration. There is no prompt field.
How is this different from WAN 2.2 Animate?
Animate drives a single character image using only the motion from a reference video and builds a new scene around it. Replace keeps the source video's existing scene, background, and camera work, and only swaps the character inside it. Both price at 2 credits per second and take the same video plus character_image inputs.