API Plan Tiers

Below are the of tiers of API usage available:

Please note that the video_translate endpoint can only be used by Scale Plan and Enterprise Plan customers.

📘

What is an API Credit?

HeyGen API Credits are consumed when creating videos, translating videos, or creating active streaming sessions with Interactive Avatars. API credits are consumed at different rates, depending on which endpoint of the HeyGen API is being used:

API functionality	API Credit costs
video/generate endpoint footage featuring a Photo Avatar	1 API credit = 5 minutes
video/generate endpoint footage featuring a Video Avatar	1 API credit = 2 minutes
video_translate endpoint footage	1 API credit = 20 seconds
Active streaming API sessions	1 API credit = 10 minutes active streaming
photo_avatar/photo/generate	0.25 API credits per call
photo_avatar/look/generate	0.25 API credits per call
photo_avatar/train	4 API credits per call
photo_avatar/add_motion	1 API credit per call
photo_avatar/add_sound_effect	Free
photo_avatar/upscale	0.5 API credits per call

You can review your API Credit consumption on the Usage History section of the API tab in your Space Settings

Video Length and Resolution Limits

The length of videos that can be translated or generated via API is limited by Tier:

Video/Generate Endpoint Limits

Resources

HeyGen limits the size and format of resources such as images, audio or video, that can be uploaded to the Video/Generate endpoint. Render errors may occur if resources are invalid.

Please ensure:

• Your resource URL is public and accessible to anyone.
• The provided asset is in one of the supported formats.
• Check that the asset file extension aligns with the actual format.
• Verify that the asset file is not corrupted or misformatted.

Avatar Input

Avatar input can be either text or audio:

• The Input Text - which is the text the Avatar speaks - should be less than 5,000 characters.
• The duration of audio input for avatars should not exceed 10 minutes (3600 seconds).

Invalid inputs will result in video render failures.

Frame Rate & Dimension

• The frame rate of video output containing avatars is 25fps.
• dimension parameter width or height must be less than 4096 and greater than 128.

Scenes

• Maximum 50 Video Input/scenes allowed per video.

Streaming endpoint

Repeat Task

• Up to 1,000 characters at a time can be sent to the Interactive Avatar to repeat text.