Get a weekly rundown of the latest AI models and research... subscribe!



The DeepFake AI model is a machine learning system designed to generate synthetic media by swapping faces in videos. The model takes two inputs: a 'swap_image' (the face to be used) and a 'target_video' (the video where the face will be applied). The output is a new video where the face from the 'swap_image' has been applied to the characters in the 'target_video'. This advanced AI technology can create highly realistic videos using deep learning techniques.

Use cases

The DeepFake AI model has a wide spectrum of applications. As implied by the name, one of its primary uses is in creating deepfakes, where the face of one person in an image or video is replaced by another person's face. For example, the input to this model could consist of a 'swap image' of a celebrity and a 'target video' of a common person. The model can output a video in which the common person's face has been convincingly swapped with the celebrity's. Beyond entertainment, this technology could find use in film and television production, reducing the need for expensive CGI or stand-ins. More practical applications, though potentially controversial, could be found in legal and security settings. Mock-up visuals could be created for crime reconstructions or simulating potential security scenarios. Furthermore, the technology could be used for educational purposes, such as historical recreations in a more engaging format. However, ethical considerations are paramount when creating deepfake videos given their potential for misuse. Other potential products that could emanate from this model include mobile apps for personalized content creation, advertising and marketing tools that offer customizable celebrity endorsements, or even digital agents for customer service that can take on the likeness of a human representative. Innovative uses in the fashion industry can also be visualized, with models replaced by celebrities or custom avatars for a personalized shopping experience. Virtual reality environments could utilize this technology for more realistic characters. In addition, more interactive video conferencing could be possible, where participants choose avatars or public figures as their on-screen representation.



Cost per run
Avg run time

Creator Models

No other models by this creator

Similar Models

Try it!

You can use this area to play around with demo applications that incorporate the Deepfake model. These demos are maintained and hosted externally by third-party creators. If you see an error, message me on Twitter.

Currently, there are no demos available for this model.


Summary of this model and related resources.

Model NameDeepfake
DeepFake AI
Model LinkView on Replicate
API SpecView on Replicate
Github LinkNo Github link provided
Paper LinkNo paper link provided


How popular is this model, by number of runs? How popular is the creator, by the sum of all their runs?

Model Rank
Creator Rank


How much does it cost to run this model? How long, on average, does it take to complete a run?

Cost per Run$-
Prediction Hardware-
Average Completion Time-