facefusion/facefusion
Face fusion model that swaps faces from source images onto a target video, producing a video with seamlessly blended facial features.
$1 / request
GPU: A100
Pricing
facefusion/
facefusion
Pricing for Synexa AI models works differently from other providers. Instead of being billed by time, you are billed by input and output, making pricing more predictable.
Output
$1.0000 / video
or
1 videos / $1
For example, generating 100 videos should cost around $100.00.
Check out our docs for more information about how per-request pricing works on Synexa.
| Provider | Price ($) | Saving (%) |
|---|---|---|
| Synexa | $1.0000 | - |
| replicate | $2.0000 | 50.0% |
Readme
Given source face images and a target video, the face fusion model automatically detects and replaces faces in the video with faces from the source images. The result is a video where the swapped faces blend naturally with the target video's appearance and lighting.
Limitations:
- Face contours should be complete without significant occlusion
- Supports face lateralization up to approximately 30 degrees
- Face size should be at least 64x64 pixels
- Face area should not exceed 2/3 of the frame
- Large differences in face shape between source and target may affect edge blending