curl --request POST \
  --url https://api.evolink.ai/v1/videos/generations \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "minimax-h3-text-to-video",
  "prompt": "Epic space-opera trailer: a female captain stands alone before a vast observation window as the last fleet assembles outside. The fleet jumps in a blinding white flash and the bridge shakes violently; when the light fades, she is left alone in silent deep space. Cinematic lighting, slow push-in, grand yet restrained atmosphere.",
  "duration": 4,
  "quality": "768p",
  "aspect_ratio": "16:9"
}
'

{
  "created": 1761313744,
  "id": "task-unified-1774857405-abc123",
  "model": "minimax-h3-text-to-video",
  "object": "video.generation.task",
  "progress": 0,
  "status": "pending",
  "task_info": {
    "can_cancel": false,
    "estimated_time": 165,
    "video_duration": 4
  },
  "type": "video",
  "usage": {
    "billing_rule": "per_second",
    "credits_reserved": 44.2,
    "user_group": "default"
  }
}

{
  "error": {
    "code": "invalid_request",
    "message": "Invalid request parameters",
    "type": "invalid_request_error"
  }
}

{
  "error": {
    "code": "unauthorized",
    "message": "Invalid or expired token",
    "type": "authentication_error"
  }
}

{
  "error": {
    "code": "insufficient_quota",
    "message": "Insufficient quota. Please top up your account.",
    "type": "insufficient_quota"
  }
}

{
  "error": {
    "code": "model_access_denied",
    "message": "Token does not have access to the specified model",
    "type": "invalid_request_error"
  }
}

{
  "error": {
    "code": "rate_limit_exceeded",
    "message": "Too many requests, please try again later",
    "type": "rate_limit_error"
  }
}

{
  "error": {
    "code": "internal_error",
    "message": "Internal server error",
    "type": "api_error"
  }
}

Minimax H3

Minimax H3 / H3 Max Complete Parameter Guide

Unified API for the entire Minimax H3 and H3 Max series. Select the generation mode with the model parameter.

3 generation modes, 5 models:

Mode	Model ID	Input assets
Text-to-Video	`minimax-h3-text-to-video`	`prompt` only
Text-to-Video	`minimax-h3-max-text-to-video`	`prompt` only
Image-to-Video	`minimax-h3-image-to-video`	`prompt` + `image_start` and/or `image_end`
Image-to-Video	`minimax-h3-max-image-to-video`	`prompt` + `image_start` and/or `image_end`
Reference-to-Video	`minimax-h3-reference-to-video`	`prompt` + `image_urls` / `video_urls` / `audio_urls`

Parameter differences:

Minimax H3: output duration from 4 to 15 seconds; resolution 768p (default) or 2k
Minimax H3 Max: output duration from 5 to 15 seconds; resolution 480p or 768p (default)

Shared capabilities:

Asynchronous processing; use the returned task ID to query the task
Minimax H3 and H3 Max tasks cannot be canceled
Generated video URLs remain valid for 24 hours; save the result promptly

POST

videos

generations

curl --request POST \
  --url https://api.evolink.ai/v1/videos/generations \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "minimax-h3-text-to-video",
  "prompt": "Epic space-opera trailer: a female captain stands alone before a vast observation window as the last fleet assembles outside. The fleet jumps in a blinding white flash and the bridge shakes violently; when the light fades, she is left alone in silent deep space. Cinematic lighting, slow push-in, grand yet restrained atmosphere.",
  "duration": 4,
  "quality": "768p",
  "aspect_ratio": "16:9"
}
'

{
  "created": 1761313744,
  "id": "task-unified-1774857405-abc123",
  "model": "minimax-h3-text-to-video",
  "object": "video.generation.task",
  "progress": 0,
  "status": "pending",
  "task_info": {
    "can_cancel": false,
    "estimated_time": 165,
    "video_duration": 4
  },
  "type": "video",
  "usage": {
    "billing_rule": "per_second",
    "credits_reserved": 44.2,
    "user_group": "default"
  }
}

{
  "error": {
    "code": "invalid_request",
    "message": "Invalid request parameters",
    "type": "invalid_request_error"
  }
}

{
  "error": {
    "code": "unauthorized",
    "message": "Invalid or expired token",
    "type": "authentication_error"
  }
}

{
  "error": {
    "code": "insufficient_quota",
    "message": "Insufficient quota. Please top up your account.",
    "type": "insufficient_quota"
  }
}

{
  "error": {
    "code": "model_access_denied",
    "message": "Token does not have access to the specified model",
    "type": "invalid_request_error"
  }
}

{
  "error": {
    "code": "rate_limit_exceeded",
    "message": "Too many requests, please try again later",
    "type": "rate_limit_error"
  }
}

{
  "error": {
    "code": "internal_error",
    "message": "Internal server error",
    "type": "api_error"
  }
}

Authorizations

Authorization

string

header

required

##All endpoints require Bearer Token authentication##

Get API Key:

Visit the API Key Management Page to obtain your API Key

Add to request header:

Authorization: Bearer YOUR_API_KEY

Body

application/json

model

enum<string>

required

Video generation model name.

Model ID	Mode	Asset parameters
`minimax-h3-text-to-video`	Text-to-video	Does not accept media assets
`minimax-h3-max-text-to-video`	Text-to-video	Does not accept media assets
`minimax-h3-image-to-video`	Image-to-video	`image_start` and/or `image_end`
`minimax-h3-max-image-to-video`	Image-to-video	`image_start` and/or `image_end`
`minimax-h3-reference-to-video`	Multimodal reference-to-video	`image_urls` / `video_urls` / `audio_urls`

Available options:

minimax-h3-text-to-video,

minimax-h3-max-text-to-video,

minimax-h3-image-to-video,

minimax-h3-max-image-to-video,

minimax-h3-reference-to-video

Example:

"minimax-h3-text-to-video"

prompt

string

required

Describe the video you want to generate. This field is required and cannot be empty.

Prompt requirements:

Chinese and English are supported
Maximum length 7000 characters (Chinese and English are both counted by character)
The model may ignore some details in an overly long prompt

Usage by model:

Text-to-Video: Describe the scene, subject, action, camera, and mood. Do not provide any media fields
Image-to-Video: Describe how subjects in the first/last frame should move and how the camera should change; use image_start and/or image_end
Reference-to-Video: Describe how to use the reference assets. Refer to them as Image 1, Image 2, Video 1, Audio 1, and so on; do not use Seedance-style names such as @image1 or @video1. Numbering starts at 1 and follows the order of each URL array

Required string length: 1 - 7000

Example:

"Cinematic scene: an astronaut stands before the observation window of a lunar base as the camera slowly pushes in and Earth rises above the distant horizon."

image_start

string<uri>

Applicable models:

minimax-h3-image-to-video
minimax-h3-max-image-to-video
Text-to-video and reference-to-video do not support this parameter

HTTP(S) URL of the first-frame image.

Combination rules:

Provide at least one of image_start and image_end
image_start only: first-frame image-to-video
Both fields: first-and-last-frame image-to-video
At most 1 first frame; this field accepts one URL

Image requirements:

Formats: JPG, JPEG, PNG, WEBP, HEIC, HEIF
File size: no more than 30MB
Width and height: each from 256 to 5760 px
Aspect ratio (width/height): 0.4–2.5
The URL must use HTTP(S) and be directly accessible by the service
The complete JSON request body must not exceed 64MB; Base64 and mm_file:// inputs are not accepted

Example:

"https://cdn.hailuoai.com/prod/hailuo_demo/testsets/H3_AA_I2VA/gallery/sr_v17_variants_seed42_43_20260724/inputs/4a3a90bf9100_KDmcbkhzYo5sjjxr9FqcVmWVnzb.png"

image_end

string<uri>

Applicable models:

minimax-h3-image-to-video
minimax-h3-max-image-to-video
Text-to-video and reference-to-video do not support this parameter

HTTP(S) URL of the last-frame image.

Combination rules:

Provide at least one of image_end and image_start
image_end only: the model generates natural motion that ends on this frame
Both fields: first-and-last-frame image-to-video
At most 1 last frame; this field accepts one URL

Image requirements:

Formats: JPG, JPEG, PNG, WEBP, HEIC, HEIF
File size: no more than 30MB
Width and height: each from 256 to 5760 px
Aspect ratio (width/height): 0.4–2.5
The URL must use HTTP(S) and be directly accessible by the service
The complete JSON request body must not exceed 64MB; Base64 and mm_file:// inputs are not accepted

Example:

"https://images.unsplash.com/photo-1515003197210-e0cd71810b5f?auto=format&fit=crop&w=1920&q=85"

image_urls

string<uri>[]

Applicable models:

Only minimax-h3-reference-to-video
Text-to-video and image-to-video do not support this parameter

Array of HTTP(S) reference-image URLs. Defaults to []; maximum 9 images.

Role and numbering:

Every image is used as reference_image
The first item is Image 1, the second is Image 2, and so on

Image requirements:

Formats: JPG, JPEG, PNG, WEBP, HEIC, HEIF
File size: no more than 30MB each
Width and height: each from 256 to 5760 px
Aspect ratio (width/height): 0.4–2.5
URLs must use HTTP(S) and be directly accessible by the service
The complete JSON request body must not exceed 64MB; Base64 and mm_file:// inputs are not accepted

Combination rules:

Provide at least one of image_urls, video_urls, or audio_urls
The three reference types combined must not exceed 12 items (images + videos + audio clips)
Leaving all three empty returns a parameter error; use minimax-h3-text-to-video or minimax-h3-max-text-to-video for text-only generation

Maximum array length: 9

Example:

[
  "https://cdn.hailuoai.com/prod/hailuo_demo/testsets/H3_AA_I2VA/gallery/sr_v17_variants_seed42_43_20260724/inputs/4a3a90bf9100_KDmcbkhzYo5sjjxr9FqcVmWVnzb.png"
]

video_urls

string<uri>[]

Applicable models:

Only minimax-h3-reference-to-video
Text-to-video and image-to-video do not support this parameter

Array of HTTP(S) reference-video URLs. Defaults to []; maximum 3 videos.

Role and numbering:

Every video is used as reference_video
The first item is Video 1, the second is Video 2, and so on

Video requirements:

Containers: MP4 (.mp4), MOV (.mov)
Video codecs: H.264/AVC, H.265/HEVC
Embedded audio codecs: AAC, MP3
File size: no more than 50MB each
Duration: 2–15 seconds per clip; total reference-video duration no more than 15 seconds
Width and height: each from 256 to 5760 px
Aspect ratio (width/height): 0.4–2.5
Frame rate: 23.976–60 FPS
URLs must use HTTP(S) and be directly accessible by the service
The complete JSON request body must not exceed 64MB; Base64 and mm_file:// inputs are not accepted

Billing:

Reference-video input duration is billable

Combination rules:

Provide at least one of image_urls, video_urls, or audio_urls
The three reference types combined must not exceed 12 items (images + videos + audio clips)
Leaving all three empty returns a parameter error; use minimax-h3-text-to-video or minimax-h3-max-text-to-video for text-only generation

Maximum array length: 3

Example:

[
  "https://cdn.hailuoai.com/prod/hailuo_demo/testsets/h3_promo_eval_ref2va/gallery/sr_v2p26_trio_seed42_20260724/inputs/297573323635_00_%E8%A7%86%E9%A2%911_YnyRbxEwio_video_20260525_163755_1927e9d3.mp4"
]

audio_urls

string<uri>[]

Applicable models:

Only minimax-h3-reference-to-video
Text-to-video and image-to-video do not support this parameter

Array of HTTP(S) reference-audio URLs. Defaults to []; maximum 3 clips.

Role and numbering:

Every audio clip is used as reference_audio
The first item is Audio 1, the second is Audio 2, and so on

Audio requirements:

Formats: WAV, MP3
File size: no more than 15MB each
Duration: 2–15 seconds per clip; total reference-audio duration no more than 15 seconds
URLs must use HTTP(S) and be directly accessible by the service
The complete JSON request body must not exceed 64MB; Base64 and mm_file:// inputs are not accepted

Billing:

Reference-audio duration is not billable

Combination rules:

audio_urls can be used on its own; no reference image or video is required
The three reference types combined must not exceed 12 items (images + videos + audio clips)
Leaving all three empty returns a parameter error; use minimax-h3-text-to-video or minimax-h3-max-text-to-video for text-only generation

Maximum array length: 3

Example:

[
  "https://cdn.hailuoai.com/prod/hailuo_demo/testsets/h3_promo_eval_ref2va/gallery/sr_v2p26_trio_seed42_20260724/inputs/f463d523c5ce_01_%E9%9F%B3%E9%A2%911_RSLcbpzJPo_6%E6%9C%885%E6%97%A5(1).mp3"
]

duration

integer

Output video duration in seconds. The value range and default depend on model.

Applicable models:

All 5 Minimax H3 and H3 Max models

Model restrictions:

Minimax H3: integers from 4 through 15, inclusive; defaults to 4 seconds
Minimax H3 Max: integers from 5 through 15, inclusive; defaults to 5 seconds
Decimals, numeric strings, auto, and -1 are not supported
Output duration directly affects billing

Required range: 4 <= x <= 15

Example:

5

quality

enum<string>

default:768p

Output video resolution. Defaults to 768p.

Applicable models:

All 5 Minimax H3 and H3 Max models

Available values by model:

Minimax H3: 768p (default), 2k
Minimax H3 Max: 480p, 768p (default)

Notes:

Omitting this parameter produces 768p output
Output bitrate, frame rate, video codec, and audio codec are selected by the platform and are not configurable

Available options:

480p,

768p,

2k

Example:

"768p"

aspect_ratio

enum<string>

Output video aspect ratio.

Applicable models and defaults:

minimax-h3-text-to-video, minimax-h3-max-text-to-video: explicit aspect ratios only; defaults to 16:9
minimax-h3-reference-to-video: supports all options below; defaults to adaptive
minimax-h3-image-to-video, minimax-h3-max-image-to-video: the model automatically adapts the output aspect ratio to the input image; do not send this parameter in image-to-video requests

Available values:

adaptive: reference-to-video only; the model selects a suitable ratio from the reference assets and prompt
16:9: Landscape
21:9: Ultrawide
4:3: Standard landscape
1:1: Square
3:4: Standard portrait
9:16: Portrait

Available options:

adaptive,

16:9,

21:9,

4:3,

1:1,

3:4,

9:16

Example:

"16:9"

callback_url

string<uri>

HTTPS callback URL for task completion

Callback timing:

Triggered when the task is completed or failed
Sent after billing confirmation is complete

Security restrictions:

Only HTTPS protocol is supported
Callbacks to private IP addresses are prohibited (127.0.0.1, 10.x.x.x, 172.16-31.x.x, 192.168.x.x, etc.)
URL length must not exceed 2048 characters

Callback mechanism:

Timeout: 10 seconds
Up to 3 retries after failure (at 1/2/4 seconds after failure respectively)
Callback response body format is consistent with the task query endpoint response format
A 2xx status code is considered successful; other status codes trigger retries

Applicable models:

All 5 Minimax H3 and H3 Max models

Pattern: ^https://

Example:

"https://your-domain.com/webhooks/video-task-completed"

Response

Video generation task created successfully

created

integer

Task creation timestamp

Example:

1761313744

string

Task ID

Example:

"task-unified-1774857405-abc123"

model

string

Actual model name used

Example:

"minimax-h3-text-to-video"

object

enum<string>

Specific type of the task

Available options:

video.generation.task

progress

integer

Task progress percentage (0-100)

Required range: 0 <= x <= 100

Example:

0

status

enum<string>

Task status

Available options:

pending,

processing,

completed,

failed

Example:

"pending"

task_info

object

Video task details

Show child attributes

type

enum<string>

Output type of the task

Available options:

text,

image,

audio,

video

Example:

"video"

usage

object

Usage and billing information

Show child attributes

Image Series

Video Series

Audio Series

Text Series

Account Management

Task Management

File Management

Minimax H3 / H3 Max Complete Parameter Guide

Authorizations

Body

Response