HappyHorse 1.1 R2V

HappyHorse 1.1 R2V generates a video with synchronized audio from a prompt and 1 to 9 reference images. Use it when people, products, scenes, or visual details must remain recognizable across shots. R2V means reference-to-video. Reference images guide identity and appearance; they are not treated as the first frame.

Key capabilities

Multi-image reference - Supply 1 to 9 images
Reference-aware prompts - Address each image with [Image 1], [Image 2], and so on
Synchronized audio - Generate audio with the video
Resolution - 720P or 1080P
Duration - An integer from 3 to 15 seconds
Aspect ratio control - Nine supported landscape, portrait, square, and ultra-wide ratios

Reference image requirements

Formats: JPEG, JPG, PNG, or WEBP
Input: public HTTP/HTTPS URL or Base64 data URL
Count: 1 to 9 images
Resolution: shortest side at least 400 pixels; 720P or clearer is recommended
File size: up to 20 MB per image

Quick example

cURL

curl https://www.anyfast.ai/v1/video/generations \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -H 'Content-Type: application/json' \
  -d '{
    "model": "happyhorse-1.1-r2v",
    "input": {
      "prompt": "The woman in [Image 1] enters the room and opens the folding fan from [Image 2]. The camera moves into a close-up.",
      "media": [
        {
          "type": "reference_image",
          "url": "https://example.com/person.jpg"
        },
        {
          "type": "reference_image",
          "url": "https://example.com/fan.jpg"
        }
      ]
    },
    "parameters": {
      "resolution": "720P",
      "ratio": "16:9",
      "duration": 5,
      "watermark": false
    }
  }'

input.media order defines image numbering. The first item is [Image 1], the second is [Image 2], and so on. Name the specific subject or object when referring to an image in the prompt.

Parameters

Parameter	Type	Required	Description
`model`	string	Yes	Must be `happyhorse-1.1-r2v`
`input.prompt`	string	Yes	Video description with optional `[Image N]` references. Maximum 5,000 non-Chinese characters or 2,500 Chinese characters
`input.media`	array	Yes	1 to 9 reference image objects
`input.media[].type`	string	Yes	Must be `reference_image`
`input.media[].url`	string	Yes	Public image URL or Base64 data URL
`parameters.resolution`	string	No	`720P` or `1080P`. Default: `1080P`
`parameters.ratio`	string	No	`16:9`, `9:16`, `3:4`, `4:3`, `4:5`, `5:4`, `1:1`, `9:21`, or `21:9`. Default: `16:9`
`parameters.duration`	integer	No	Video duration from 3 to 15 seconds. Default: `5`
`parameters.watermark`	boolean	No	Whether to add the HappyHorse watermark. Default: `true`
`parameters.seed`	integer	No	Random seed in `[0, 2147483647]`

API Reference

View the request and response schema.

Task Query

Poll task status and retrieve the result URL.

Official reference

See the Alibaba Cloud Model Studio HappyHorse reference-to-video API reference.

Getting started

Model Introduction

Resources

HappyHorse 1.1 R2V

Key capabilities

Reference image requirements

Quick example

Parameters

API Reference

Task Query

Official reference

​Key capabilities

​Reference image requirements

​Quick example

​Parameters

API Reference

Task Query

​Official reference

Key capabilities

Reference image requirements

Quick example

Parameters

Official reference