Hailuo 3 Reference To Video

curl -X POST https://api.vibetool.ai/v1/videos/generations \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "minimax-h3-reference-to-video",
    "prompt": "the model from the reference photos walks through a sunlit atrium wearing the featured jacket",
    "image_urls": [
      "https://example.com/character-ref.png",
      "https://example.com/product-ref.png"
    ],
    "duration": 6
  }'

import requests

url = "https://api.vibetool.ai/v1/videos/generations"
headers = {
    "Authorization": "Bearer YOUR_API_KEY",
    "Content-Type": "application/json"
}
payload = {
    "model": "minimax-h3-reference-to-video",
    "prompt": "the model from the reference photos walks through a sunlit atrium wearing the featured jacket",
    "image_urls": [
        "https://example.com/character-ref.png",
        "https://example.com/product-ref.png"
    ],
    "duration": 6
}

r = requests.post(url, json=payload, headers=headers, timeout=60)
print(r.status_code, r.json())

const response = await fetch("https://api.vibetool.ai/v1/videos/generations", {
  method: "POST",
  headers: {
    "Authorization": "Bearer YOUR_API_KEY",
    "Content-Type": "application/json"
  },
  body: JSON.stringify({
    model: "minimax-h3-reference-to-video",
    prompt: "the model from the reference photos walks through a sunlit atrium wearing the featured jacket",
    image_urls: [
      "https://example.com/character-ref.png",
      "https://example.com/product-ref.png"
    ],
    duration: 6
  })
});
const data = await response.json();
console.log(data);

package main

import (
	"bytes"
	"encoding/json"
	"fmt"
	"net/http"
)

func main() {
	payload := map[string]interface{}{
		"model":  "minimax-h3-reference-to-video",
		"prompt": "the model from the reference photos walks through a sunlit atrium wearing the featured jacket",
		"image_urls": []string{
			"https://example.com/character-ref.png",
			"https://example.com/product-ref.png",
		},
		"duration": 6,
	}
	b, _ := json.Marshal(payload)

	req, _ := http.NewRequest("POST", "https://api.vibetool.ai/v1/videos/generations", bytes.NewReader(b))
	req.Header.Set("Authorization", "Bearer YOUR_API_KEY")
	req.Header.Set("Content-Type", "application/json")

	client := &http.Client{}
	resp, err := client.Do(req)
	if err != nil {
		panic(err)
	}
	defer resp.Body.Close()

	var result map[string]interface{}
	json.NewDecoder(resp.Body).Decode(&result)
	fmt.Println(result)
}

{
  "id": "gen_vid_h3_003",
  "object": "video.generation",
  "model": "minimax-h3-reference-to-video",
  "created_at": 1786549095,
  "status": "pending",
  "task_id": "vid_202608130145_1c2d3e4f"
}

{
  "error": {
    "message": "The prompt field is required",
    "type": "invalid_request_error",
    "param": "prompt",
    "code": 400
  }
}

POST

videos

generations

curl -X POST https://api.vibetool.ai/v1/videos/generations \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "minimax-h3-reference-to-video",
    "prompt": "the model from the reference photos walks through a sunlit atrium wearing the featured jacket",
    "image_urls": [
      "https://example.com/character-ref.png",
      "https://example.com/product-ref.png"
    ],
    "duration": 6
  }'

import requests

url = "https://api.vibetool.ai/v1/videos/generations"
headers = {
    "Authorization": "Bearer YOUR_API_KEY",
    "Content-Type": "application/json"
}
payload = {
    "model": "minimax-h3-reference-to-video",
    "prompt": "the model from the reference photos walks through a sunlit atrium wearing the featured jacket",
    "image_urls": [
        "https://example.com/character-ref.png",
        "https://example.com/product-ref.png"
    ],
    "duration": 6
}

r = requests.post(url, json=payload, headers=headers, timeout=60)
print(r.status_code, r.json())

const response = await fetch("https://api.vibetool.ai/v1/videos/generations", {
  method: "POST",
  headers: {
    "Authorization": "Bearer YOUR_API_KEY",
    "Content-Type": "application/json"
  },
  body: JSON.stringify({
    model: "minimax-h3-reference-to-video",
    prompt: "the model from the reference photos walks through a sunlit atrium wearing the featured jacket",
    image_urls: [
      "https://example.com/character-ref.png",
      "https://example.com/product-ref.png"
    ],
    duration: 6
  })
});
const data = await response.json();
console.log(data);

package main

import (
	"bytes"
	"encoding/json"
	"fmt"
	"net/http"
)

func main() {
	payload := map[string]interface{}{
		"model":  "minimax-h3-reference-to-video",
		"prompt": "the model from the reference photos walks through a sunlit atrium wearing the featured jacket",
		"image_urls": []string{
			"https://example.com/character-ref.png",
			"https://example.com/product-ref.png",
		},
		"duration": 6,
	}
	b, _ := json.Marshal(payload)

	req, _ := http.NewRequest("POST", "https://api.vibetool.ai/v1/videos/generations", bytes.NewReader(b))
	req.Header.Set("Authorization", "Bearer YOUR_API_KEY")
	req.Header.Set("Content-Type", "application/json")

	client := &http.Client{}
	resp, err := client.Do(req)
	if err != nil {
		panic(err)
	}
	defer resp.Body.Close()

	var result map[string]interface{}
	json.NewDecoder(resp.Body).Decode(&result)
	fmt.Println(result)
}

{
  "id": "gen_vid_h3_003",
  "object": "video.generation",
  "model": "minimax-h3-reference-to-video",
  "created_at": 1786549095,
  "status": "pending",
  "task_id": "vid_202608130145_1c2d3e4f"
}

{
  "error": {
    "message": "The prompt field is required",
    "type": "invalid_request_error",
    "param": "prompt",
    "code": 400
  }
}

Pricing: See vibetool.ai/pricing for current rates.

Hailuo 3 (MiniMax H3) is MiniMax’s flagship video generation model, exposed through Vibetool as a reference-to-video endpoint. Supply reference images, video clips, or audio tracks alongside your prompt, and the model keeps characters, products, and visual style consistent with those references in the generated 2K output — ideal for advertising, e-commerce, and narrative work.

Input Modalities

Text – a prompt describing the desired video
Image – up to 9 reference images via image_urls
Video – up to 3 reference clips via video_urls (each 2–15 seconds)
Audio – up to 3 reference audio tracks via audio_url (each 2–15 seconds)

Output Modality

Video – a 2K MP4 file, 4–15 seconds long

Authorization

string

required

All APIs require Bearer Token authentication. Format: Bearer YOUR_API_KEY

Request Body

string

required

The model identifier. Use "minimax-h3-reference-to-video".

string

required

Text description of the video to generate. Reference the supplied assets to keep characters and products consistent.

array of strings

Up to 9 publicly accessible reference image URLs for character, product, or style consistency.

array of strings

Up to 3 publicly accessible reference video URLs, each 2–15 seconds long.

string

A publicly accessible reference audio URL, 2–15 seconds long.

integer

default:"4"

Video duration in seconds — any integer from 4 to 15. Defaults to 4 when omitted.

Billing is per second of output video. When a video reference is supplied, the reference clip’s length is added to the billable duration; image and audio references bill on output seconds only. Additional charges apply to input images beyond the first five — see the pricing page. One video per request; the n parameter is not supported.

Response Schema

string

required

Unique identifier for the generation task.

string

required

Type of object, always "video.generation".

string

required

The model used, e.g., "minimax-h3-reference-to-video".

integer

required

Unix timestamp (seconds) of task creation.

string

required

Initial status: "pending" or "processing". Use the Get Video Status endpoint to poll for completion.

string

required

The task ID to query in the status endpoint.

curl -X POST https://api.vibetool.ai/v1/videos/generations \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "minimax-h3-reference-to-video",
    "prompt": "the model from the reference photos walks through a sunlit atrium wearing the featured jacket",
    "image_urls": [
      "https://example.com/character-ref.png",
      "https://example.com/product-ref.png"
    ],
    "duration": 6
  }'

import requests

url = "https://api.vibetool.ai/v1/videos/generations"
headers = {
    "Authorization": "Bearer YOUR_API_KEY",
    "Content-Type": "application/json"
}
payload = {
    "model": "minimax-h3-reference-to-video",
    "prompt": "the model from the reference photos walks through a sunlit atrium wearing the featured jacket",
    "image_urls": [
        "https://example.com/character-ref.png",
        "https://example.com/product-ref.png"
    ],
    "duration": 6
}

r = requests.post(url, json=payload, headers=headers, timeout=60)
print(r.status_code, r.json())

const response = await fetch("https://api.vibetool.ai/v1/videos/generations", {
  method: "POST",
  headers: {
    "Authorization": "Bearer YOUR_API_KEY",
    "Content-Type": "application/json"
  },
  body: JSON.stringify({
    model: "minimax-h3-reference-to-video",
    prompt: "the model from the reference photos walks through a sunlit atrium wearing the featured jacket",
    image_urls: [
      "https://example.com/character-ref.png",
      "https://example.com/product-ref.png"
    ],
    duration: 6
  })
});
const data = await response.json();
console.log(data);

package main

import (
	"bytes"
	"encoding/json"
	"fmt"
	"net/http"
)

func main() {
	payload := map[string]interface{}{
		"model":  "minimax-h3-reference-to-video",
		"prompt": "the model from the reference photos walks through a sunlit atrium wearing the featured jacket",
		"image_urls": []string{
			"https://example.com/character-ref.png",
			"https://example.com/product-ref.png",
		},
		"duration": 6,
	}
	b, _ := json.Marshal(payload)

	req, _ := http.NewRequest("POST", "https://api.vibetool.ai/v1/videos/generations", bytes.NewReader(b))
	req.Header.Set("Authorization", "Bearer YOUR_API_KEY")
	req.Header.Set("Content-Type", "application/json")

	client := &http.Client{}
	resp, err := client.Do(req)
	if err != nil {
		panic(err)
	}
	defer resp.Body.Close()

	var result map[string]interface{}
	json.NewDecoder(resp.Body).Decode(&result)
	fmt.Println(result)
}

{
  "id": "gen_vid_h3_003",
  "object": "video.generation",
  "model": "minimax-h3-reference-to-video",
  "created_at": 1786549095,
  "status": "pending",
  "task_id": "vid_202608130145_1c2d3e4f"
}

{
  "error": {
    "message": "The prompt field is required",
    "type": "invalid_request_error",
    "param": "prompt",
    "code": 400
  }
}

Once the task is created, poll the Get Video Status endpoint with the returned task_id until status is "completed" and the generated video URL is available.

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

Use minimax-h3-reference-to-video.

Available options:

minimax-h3-reference-to-video

prompt

string

required

Text description of the video to generate

image_urls

string<uri>[]

Up to 9 reference images

Maximum array length: 9

video_urls

string<uri>[]

Up to 3 reference clips, 2-15s each

Maximum array length: 3

audio_url

string<uri>

Reference audio, 2-15s

duration

integer

default:4

Duration in seconds (4-15)

Required range: 4 <= x <= 15

Response

Task created

string

Unique identifier for the generation task

object

string

Example:

"video.generation"

model

string

created_at

integer

Unix timestamp (seconds)

status

enum<string>

Initial task status

Available options:

pending,

processing

task_id

string

Task ID for the status polling endpoint

Getting Started

Text Generation

Image Generation

Video Generation

Hailuo 3 Reference To Video

Input Modalities

Output Modality

Authorization

Request Body

Response Schema

Authorizations

Body

Response

​Input Modalities

​Output Modality

​Authorization

​Request Body

​Response Schema

Authorizations

Body

Response

Input Modalities

Output Modality

Authorization

Request Body

Response Schema