Skip to main content
POST
H3 Max Turbo generates 5–15 second videos at 480p, 768p, or 1080p. Submit a task with model: "h3-max-turbo", then poll its status or provide a callback_url to receive the result.

Generation Modes

  • Text to video: omit image_urls or pass an empty array.
  • First frame: provide one image in image_urls.
  • First and last frames: provide two images in image_urls, with the starting frame first and the ending frame second.
Image-guided generation follows the first image’s aspect ratio. Omit aspect_ratio whenever image_urls contains images.

Parameter Notes

  • prompt is required in every mode and accepts up to 50,000 characters.
  • duration accepts an integer from 5 through 15 seconds and defaults to 5.
  • resolution accepts 480p, 768p, or 1080p and defaults to 768p. Uppercase P is also accepted.
  • enable_safety_checker (boolean) enables the safety checker when true. Defaults to true in every mode; set false to disable it.
  • Text-to-video supports 21:9, 16:9, 4:3, 1:1, 3:4, and 9:16. Its default aspect ratio is 16:9.
  • image_urls accepts at most two publicly accessible image URLs or Base64 images.
  • Reference media, custom audio tracks, seed, and prompt-expansion controls are not available for this model.

Billing

Video generation is billed by output duration and resolution.

Authorizations

Authorization
string
header
required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Body

application/json
model
enum<string>
required
Available options:
h3-max-turbo
Example:

"h3-max-turbo"

input
object
required
callback_url
string<uri>

Optional webhook callback URL.

Response

200 - application/json

Task submitted successfully

code
integer
required
Example:

200

data
object
required