> ## Documentation Index
> Fetch the complete documentation index at: https://docs.vidgo.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# MiniMax H3 Max Reference to Video
> Create a video from image, video and audio references.
1. Submit the request with `VIDGO_API_KEY`. A `task_id` is returned. Keep it until the task reaches `finished` or `failed`.
2. Use [Query Task Status](/api-manual/task-management/status) to retrieve the result. If you provide a `callback_url`, Vidgo sends the result to your webhook when the task finishes or fails.
This endpoint is currently unavailable. Requests return HTTP `503` without creating a task or charging credits.
## Usage Guide
Use `minimax/h3-max/reference-to-video`. When this workflow becomes available, provide at least one reference image. Optionally add reference video and audio, with at most 12 assets across all lists. Refer to assets as `Image 1`, `Video 1` and `Audio 1` in list order. Starting and ending frames cannot be combined with reference assets.
Set `model` and optional `callback_url` at the request root. Place generation parameters inside `input`.
## Parameter Details
* `prompt`: required non-empty string, up to 50,000 characters.
* `duration`: integer from 5 through 15 seconds; defaults to 5. Fractional durations are rejected.
* `resolution`: `480p`, `768p` or `1080p`; defaults to `768p`.
* `aspect_ratio`: six fixed ratios (`21:9`, `16:9`, `4:3`, `1:1`, `3:4`, `9:16`) or `adaptive`; defaults to `adaptive`.
* `reference_image_urls`: required array of 1-9 static JPEG, PNG or WebP images, at most 30 MB each. Supported shapes are `1:1`, `4:3`, `3:4`, `16:9`, `9:16`, `5:2` and `2:5`; normalize image orientation before uploading.
* `reference_video_urls`: optional array of up to 3 MP4/MOV clips, at most 50 MB each. Each clip must be 2-15 seconds, with at most 15 seconds combined. Clips must use constant 24 fps, square pixels, zero rotation metadata, and a `16:9` or `9:16` shape. After limiting frames to the requested output duration, the counted frames must be 48, 120, 240 or 360. Other timing patterns are rejected.
* `reference_audio_urls`: optional array of up to 3 MP3/WAV clips, at most 15 MB each. Each clip must be 2-15 seconds and no longer than the output duration, with at most 15 seconds combined. Audio must be mono or stereo at 32 kHz, with a decoded sample count divisible by 800.
* All reference URLs must be public HTTP(S) download links. Total assets across the three lists must not exceed 12.
## Optional parameters
* `enable_safety_checker`: boolean, defaults to `true`. Set `false` to disable it.
* `callback_url`: optional webhook URL at the request root. Query task status if you do not use a callback.
## OpenAPI
````yaml api-manual/video-series/minimax-h3-max-reference-to-video.json POST /api/generate/submit
openapi: 3.0.0
info:
title: Vidgo API - MiniMax H3 Max Reference to Video
description: Create a video from image, video and audio references.
version: 1.0.0
servers:
- url: https://api.vidgo.ai
description: Production server
security:
- BearerAuth: []
paths:
/api/generate/submit:
post:
tags:
- MiniMax H3 Max Reference to Video
summary: Submit MiniMax H3 Max Reference to Video Task
operationId: submitMinimaxH3MaxReferenceToVideoTask
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/SubmitRequest'
examples:
imageReferences:
summary: Image references (pending availability)
value:
model: minimax/h3-max/reference-to-video
input:
prompt: Image 1 walks through a sunlit street
reference_image_urls:
- https://example.com/subject.jpg
mixedReferences:
summary: Mixed references (pending availability)
value:
model: minimax/h3-max/reference-to-video
callback_url: https://your-domain.com/callback
input:
prompt: >-
Image 1 follows the movement in Video 1 with the
atmosphere of Audio 1
duration: 10
resolution: 768p
reference_image_urls:
- https://example.com/subject.jpg
reference_video_urls:
- https://example.com/motion.mp4
reference_audio_urls:
- https://example.com/atmosphere.wav
aspect_ratio: adaptive
enable_safety_checker: false
responses:
'200':
description: Task submitted successfully
content:
application/json:
schema:
$ref: '#/components/schemas/SubmitResponse'
example:
code: 200
data:
task_id: task-unified-1787970600-abcd1234
status: not_started
created_time: '2026-08-29T10:30:00Z'
'400':
description: Invalid request parameters
'401':
description: Unauthorized
'503':
description: >-
This endpoint is currently unavailable. No generation task or charge
is created.
components:
schemas:
SubmitRequest:
type: object
required:
- model
- input
additionalProperties: false
properties:
model:
type: string
example: minimax/h3-max/reference-to-video
callback_url:
type: string
format: uri
description: Optional webhook callback URL.
input:
type: object
required:
- prompt
- reference_image_urls
additionalProperties: false
properties:
prompt:
type: string
minLength: 1
maxLength: 50000
pattern: \S
description: Non-empty prompt describing the target video.
duration:
type: integer
minimum: 5
maximum: 15
default: 5
description: Requested output duration in seconds.
resolution:
type: string
enum:
- 480p
- 768p
- 1080p
default: 768p
description: Output resolution.
enable_safety_checker:
type: boolean
default: true
description: Set false to disable the safety checker.
aspect_ratio:
type: string
enum:
- '21:9'
- '16:9'
- '4:3'
- '1:1'
- '3:4'
- '9:16'
- adaptive
default: adaptive
reference_image_urls:
type: array
minItems: 1
maxItems: 9
items:
type: string
format: uri
pattern: ^https?://
description: Publicly accessible HTTP(S) download URL.
description: >-
Required reference images: static JPEG/PNG/WebP, at most 30 MB
each. Supported shapes: 1:1, 4:3, 3:4, 16:9, 9:16, 5:2, 2:5.
Normalize orientation before uploading.
reference_video_urls:
type: array
maxItems: 3
items:
type: string
format: uri
pattern: ^https?://
description: Publicly accessible HTTP(S) download URL.
description: >-
MP4/MOV, at most 50 MB each, 2-15 seconds per clip and at most
15 seconds combined. Constant 24 fps, square pixels, zero
rotation, 16:9 or 9:16. After capping at output duration,
counted frames must be 48, 120, 240 or 360.
reference_audio_urls:
type: array
maxItems: 3
items:
type: string
format: uri
pattern: ^https?://
description: Publicly accessible HTTP(S) download URL.
description: >-
MP3/WAV, at most 15 MB each, 2-15 seconds per clip and at most
15 seconds combined. Mono/stereo, 32 kHz, decoded sample count
divisible by 800. Each clip must not exceed output duration.
description: >-
At most 12 reference assets in total. Reference video and audio
require at least one reference image. This workflow is not enabled
yet.
SubmitResponse:
type: object
required:
- code
- data
properties:
code:
type: integer
example: 200
data:
type: object
required:
- task_id
- status
- created_time
properties:
task_id:
type: string
status:
type: string
enum:
- not_started
- running
- failed
created_time:
type: string
format: date-time
securitySchemes:
BearerAuth:
type: http
scheme: bearer
bearerFormat: API Key
description: 'Use `Authorization: Bearer VIDGO_API_KEY`.'
````
This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.