The HappyHorse 1 API is a practical choice for cinematic clips with smooth camera movement and expressive subject motion. On MakeInfluencer, one endpoint automatically selects text-to-video when no start frame is present and image-to-video when `startFrameUrl` is supplied.

## Live MakeInfluencer Contract

| Field           | Purpose                                                     |
| --------------- | ----------------------------------------------------------- |
| `prompt`        | Required scene, subject, motion, and camera direction       |
| `startFrameUrl` | Optional source image; switches the model to image-to-video |
| `duration`      | 3–15 seconds, default 5                                     |
| `resolution`    | `720p` or `1080p`                                           |
| `aspectRatio`   | Output composition for text-to-video requests               |

The live [HappyHorse 1 API reference](/developers/models/happyhorse-1-api) is generated from the model registry and should be your source of truth for accepted values and sample credits.

## Text-to-Video Example

Preview the request before spending:

```bash
curl --request POST \
  https://www.makeinfluencer.ai/api/v1/models/happyhorse1/preview-credits \
  --header "Authorization: Bearer $MAKEINFLUENCER_API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
    "input": {
      "prompt": "A streetwear creator crosses a rain-lit Seoul intersection, smooth low tracking shot",
      "aspectRatio": "9:16",
      "duration": 5,
      "resolution": "720p"
    }
  }'
```

Then send the same body to `/models/happyhorse1/generations` with a unique `Idempotency-Key`.

## Image-to-Video Example

Add `startFrameUrl` to animate an approved image:

```json
{
  "input": {
    "prompt": "She looks toward camera as the camera slowly pushes in; subtle coat and hair movement",
    "startFrameUrl": "https://example.com/approved-creator.jpg",
    "duration": 5,
    "resolution": "1080p"
  }
}
```

The source image controls composition, so an aspect ratio is usually unnecessary in image-to-video mode.

## When HappyHorse 1 Fits

<FeatureGrid columns="2">
  <Feature
    icon="film"
    title="Cinematic Social Clips"
    description="Use clear camera language and one focused action for polished creator content."
    index={0}
  />
  <Feature
    icon="image"
    title="Approved-Image Animation"
    description="Animate campaign stills after face, product, and styling review."
    index={1}
  />
  <Feature
    icon="globe"
    title="Landscape and Vertical"
    description="Choose aspect ratio for text generation or inherit composition from a start frame."
    index={2}
  />
  <Feature
    icon="target"
    title="Resolution-Based Budgeting"
    description="Preview 720p and 1080p separately because the per-second credit cost differs."
    index={3}
  />
</FeatureGrid>

## Prompt Formula

Use one clause for each layer:

> Subject and action + environment + camera movement + lighting + pace.

Example:

> A fashion creator steps from a black car and adjusts her sunglasses; wet city street at blue hour; smooth 35mm tracking shot; neon reflections; restrained premium-commercial pacing.

Avoid asking for several locations, outfit changes, and camera cuts inside a five-second clip. A focused action gives the model a better chance to preserve anatomy and visual continuity.

## HappyHorse 1 vs Seedance 2 Fast

Use HappyHorse when you want its cinematic motion style and are comfortable with 720p or 1080p. Evaluate [Seedance 2 Fast](/developers/models/seedance-2-fast-api) when low-resolution drafts, generated audio, an end frame, or reference images are part of the workflow.

Because both models use the same API lifecycle, a comparison only requires changing the model ID and adapting the published input fields.

## API FAQ

### Does HappyHorse support image-to-video?

Yes. Add `startFrameUrl`; the shared model adapter selects the image-to-video provider variant.

### How long can a HappyHorse clip be?

The current registry accepts durations from 3 to 15 seconds.

### Does 1080p cost more?

Yes. HappyHorse is priced per second with different 720p and 1080p rates. Use the free credit-preview endpoint for the exact MakeInfluencer quote.

### How do I retrieve the video?

Creation returns HTTP 202 and a generation ID. Poll `/api/v1/generations/video/{generationId}` until `terminal` is true. On success, `ready` is true and the response includes the video URL.


Use the HappyHorse 1 API for Alibaba text-to-video and image-to-video generation with 3–15 second clips, 720p or 1080p output, and credit previews.

HappyHorse 1 API Guide: Text and Image to Video

Create Your AI Influencer

Continue Reading

InfiniteTalk API Guide: Talking Avatar Videos from Audio

WAN 2.2 & WAN 2.6 AI Video Guide

AI Video Generation Guide

Ready to try it yourself?