Now that you have your voice, it’s time to generate some high-quality videos! You can do this by using the UI and going to the Replicas Tab, or by using the API with the Create Video Endpoint.

There are a few things to keep in mind:

  • Videos can take a few minutes to generate, depending on how long your script is
  • Tokens are used for every minute of video generated
  • You may get slightly different results everytime you generate with the same script, as the model is non-deterministic
  • You can affect how the model outputs with how you write the script. Check out Scripting for more information on this

Videos can be downloaded, or you can use the video link to share with friends and social media.

Generating Video via API


Before starting, ensure you have a Tavus account and an API key, which can be obtained from the Developer Portal.

Creating Your Video

  1. Prepare Your Script: Define the content you wish to turn into a video.
  2. Get your Replica ID: Get the ID of the Replica you’d like to use. See Replica Selection
  3. Make the API Request: Use the Create Video Endpoint to submit your video request with the
  4. Check the Response: A successful request will return details about the video generation process, including a hosted URL for the completed video.

Getting Your Video

Your video will take a few minutes to generate, depending on the length. You can check on the status of videos using the Videos section of the Developer Portal. You can also retrieve your video details using the Get Video Endpoint, including download and stream links.