JSON2Video API allows you to integrate with AI models for asset generation like images, videos, audios, etc. This page explains how to use the native integration with AI models available in JSON2Video API.
This section covers:
How to use AI models
Currently, JSON2Video API only supports AI generation for images and voiceovers, but we are working on adding videos and music generation in the short term.
Using AI models is very simple and consistant between the different AI models, making it easy to choose and switch between them.
You have two options to use the AI models:
- Using your own third-party API Keys (Bring your own API Keys, or BYOA)
- Using JSON2Video managed service
Bring your own API Keys
When using BYOA, any cost of using the AI models is charged by the third-party API provider directly to you. You are responsible for setting up your account, managing the API keys and the costs associated with the AI models and keep enough credit in your account to generate the assets you need.
Learn more about how to configure your own API keys for the AI models in Bring your own API keys mode.
JSON2Video managed service
When using JSON2Video managed service, any cost of using the AI models is charged by JSON2Video to your account. You don't need to open accounts, subscribing or providing credit card information to the third-party API providers. JSON2Video will handle all the setup and billing for you.
JSON2Video managed service is the easiest way to use the AI models, and we try to keep the price as low as possible, even lower than the public pricing of the third-party API providers.
Learn more about how pricing works for the managed service in Credit consumption.
Technical implementation
The asset generation with AI is done using the regular Elements API. The only difference is that you need to provide the AI model you want to use and the parameters for the AI model.
AI image generation
Learn more about how to generate images with AI models in AI image generation.
Published on January 13th, 2025
