Skip to main content
POST
MiniMax H3 video generation
text_to_video 不接受素材;first_frame 要求恰好一张图片,不接受视频或音频。video_multi_ref 最多接受9张图片、3段视频和3段音频,合计不超过12项,至少包含一张图片或一段视频;每类素材顺序会保留。每类素材仅使用一种字段写法。旧720p必须明确改选768P。上游完成但产物不可用时继续重试;明确失败或取消优先于旧产物。查询状态不负责结算。价格以当前目录为准,不承诺固定出片时延。 Query Video Task · Pricing

授权

Authorization
string
header
必填

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

请求体

application/json
model
enum<string>
必填
可用选项:
minimax/h3-max
prompt
string
必填
Required string length: 1 - 50000
operation
enum<string>

Must agree with media: text uses none; first_frame requires one image only.

可用选项:
text_to_video,
first_frame,
video_multi_ref
image_url
string<uri>

HTTP(S) first-frame image; do not combine with image_urls.

image_urls
string<uri>[]

Ordered images; one field spelling per media kind.

Maximum array length: 9
resolution
enum<string>
默认值:768P

Case/whitespace normalized. 720p is rejected; select 768P explicitly.

可用选项:
480P,
768P,
1080P
duration
默认值:5
可用选项:
5,
10
aspect_ratio
enum<string>
默认值:16:9
可用选项:
16:9,
9:16,
1:1
prompt_optimizer
boolean
默认值:true
video_urls
string<uri>[]

Ordered references. Total <=12, at least one image or video.

Maximum array length: 3
audio_urls
string<uri>[]

Ordered references. Total <=12, at least one image or video.

Maximum array length: 3
image_with_roles
object[]

First-frame role form. Do not combine with image_url/image_urls; not accepted by video_multi_ref.

Required array length: 1 element

响应

Queued video task: id, task_id, model, status=queued, created

id
string
task_id
string
model
string
object
enum<string>
可用选项:
video
status
enum<string>
可用选项:
queued
created
integer