Video（Based on Subject Reference）

POST

/minimaxi/v1/video_generation

Specification
Video model from Minimax, official document: https://platform.minimaxi.com/document/video_generation?key=66d1439376e52fcee2853049#whYrra7nHOP2eZy9yMMbnnU6

Price：0.7 PTC / call

Request

Header Params

Authorization

string

optional

Example:

Bearer {{YOUR_API_KEY}}

Content-Type

string

optional

Example:

application/json

Body Params application/json

model

string

required

S2V-01

prompt

string

optional

Describe the video. (Note: less than 200 words)

prompt_optimizer

boolean

optional

The default value is True, and the model will automatically optimize the input prompt to improve generation quality. If more precise control is needed, you can set this parameter to False, and the model will adhere more strictly to the instructions. In this case, it is recommended to provide a finer prompt for optimal results.

subject_reference

array [object {2}]

required

This parameter is only available when the model is selected as S2V-01. The model will generate the video based on the subject uploaded in this parameter. Currently, only single subject reference (array length of 1) is supported.

type

string

required

Main type, currently only supports "character", that is, the main character of the face.

image

string

required

The data of the main reference image: image/jpeg; base64, a Base64 encoded string in {data} format, or a URL accessible over the public network, stored in an array in string form

Example

{
  "model": "S2V-01",
  "prompt": "a man is walking on the new york street",
  "subject_reference": [
    {
      "type": "character",
      "image": [
        "https://file.302ai.cn/gpt/imgs/20241012/813d3e2cecf146b1b5a3d1e742ee3e5d.png"
      ]
    }
  ]
}

Request samples

Shell

JavaScript

Java

Swift

PHP

Python

HTTP

Objective-C

Ruby

OCaml

Dart

curl --location --request POST 'https://api.302.ai/minimaxi/v1/video_generation' \
--header 'Authorization: Bearer sk-jls4AaVBGoe1GwZD64qZA1qyKTN1MPHa4NmvH1cT68z7K1Zz' \
--header 'Content-Type: application/json' \
--data-raw '{
    "model": "S2V-01",
    "prompt": "a man is walking on the new york street",
    "subject_reference": [
        {
            "type": "character",
            "image": [
                "https://file.302ai.cn/gpt/imgs/20241012/813d3e2cecf146b1b5a3d1e742ee3e5d.png"
            ]
        }
    ]
}'

Responses

🟢200OK

application/json

Body

task_id

string

required

Used to query the task ID

base_resp

object

required

status_code

integer

required

status_msg

string

required

Example

{
  "task_id": "227311208538261",
  "base_resp": {
    "status_code": 0,
    "status_msg": "success"
  }
}

Video（Image-to-video）

Video（Camera movement control）