How to Use the HappyHorse API on GPTImage2

HappyHorse API Guide: Call HappyHorse 1.0 Through GPTImage2

HappyHorse 1.0 is now callable on GPTImage2 through the unified video API. If your search is "HappyHorse API", the fastest path is the HappyHorse API page, where you can test the model, check the current route pricing, and inspect the live API surface.

This guide explains the developer workflow so the model page can keep owning the transactional query while this article covers integration detail.

Fast Answer

Use POST /v1/videos/generations to create a HappyHorse video task.
Use GET /v1/tasks/{task_id} to poll the async result.
Available route variants in the current codebase are happyhorse-1.0-text-to-video, happyhorse-1.0-image-to-video, happyhorse-1.0-reference-to-video, and happyhorse-1.0-video-edit.
GPTImage2 route pricing is per second. The current pricing table uses 12.15 credits/s for 720p and a 1.778x multiplier for 1080p.
For API access, pricing, and playground testing, use the HappyHorse API page.

HappyHorse API Model Names

Workflow	Model name	Best for
Text to video	`happyhorse-1.0-text-to-video`	Prompt-first video generation
Image to video	`happyhorse-1.0-image-to-video`	Animating a first-frame image
Reference to video	`happyhorse-1.0-reference-to-video`	Using 1-9 reference images
Video edit	`happyhorse-1.0-video-edit`	Editing a source video with a text instruction

These variants share the same high-level GPTImage2 pattern: submit an async generation task, receive a task ID, then fetch the result when the task completes.

Basic Request Pattern

curl --request POST \
  --url https://api.gptimage2/v1/videos/generations \
  --header 'Authorization: Bearer <GPTIMAGE2_API_KEY>' \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "happyhorse-1.0-text-to-video",
    "prompt": "A cinematic product shot of a chrome horse sculpture rotating on a black glass table, studio lighting, slow camera push-in",
    "quality": "720p",
    "duration": 5,
    "aspect_ratio": "16:9"
  }'

The response returns a task ID. Use that task ID to poll the task endpoint:

curl --request GET \
  --url https://api.gptimage2/v1/tasks/<TASK_ID> \
  --header 'Authorization: Bearer <GPTIMAGE2_API_KEY>'

Parameters to Check First

Parameter	Typical values	Notes
`model`	`happyhorse-1.0-text-to-video`	Choose the HappyHorse variant for your workflow
`prompt`	Text prompt	Required for prompt-driven generation and editing
`quality`	`720p`, `1080p`	1080p uses a higher multiplier
`duration`	`3-15` seconds	Applies to generation variants, not video edit
`aspect_ratio`	`16:9`, `9:16`, `1:1`, `4:3`, `3:4`	Text-to-video and reference-to-video support aspect ratio selection
`image_urls`	1 image or 1-9 references	Required for image/reference workflows
`video_urls`	Source video URL	Required for video edit

When to Use Each HappyHorse Variant

Use text-to-video when you want the cleanest first test of HappyHorse quality. It is the best match for broad happyhorse api and happyhorse text to video searches.

Use image-to-video when the starting frame matters. This is the better fit for product visuals, character consistency, and design-driven creative tests.

Use reference-to-video when you have multiple reference images and want the prompt to refer to them as characters or objects.

Use video edit when you already have a source video and want to modify it with an instruction. Pricing for video edit is based on total duration rules rather than a simple output-duration-only estimate.

Pricing Scope

This guide discusses GPTImage2 route pricing, not a universal official vendor price.

The current HappyHorse pricing implementation uses:

Quality	Pricing basis
720p	`12.15 credits/s`
1080p	`1.778x` the 720p rate

Video edit uses a special duration rule: billing is based on input video duration plus output video duration, with source video duration capped for billing. Always verify the live price table on the HappyHorse API page before production use.

Where This Page Fits in the SEO Cluster

Search query	Best destination
`happyhorse api`	HappyHorse API page
`how to use happyhorse api`	This guide
`happyhorse pricing`	HappyHorse pricing guide
`happyhorse vs seedance`	HappyHorse vs Seedance
`happyhorse open source`	HappyHorse release watch

Open the HappyHorse API page

FAQ

Is the HappyHorse API available now?

Yes. HappyHorse 1.0 is callable on GPTImage2 now through the unified video API.

Which endpoint do I use for HappyHorse?

Use POST /v1/videos/generations to create a task and GET /v1/tasks/{task_id} to retrieve task status and results.

Which HappyHorse model name should I use?

Start with happyhorse-1.0-text-to-video for prompt-first testing. Use happyhorse-1.0-image-to-video, happyhorse-1.0-reference-to-video, or happyhorse-1.0-video-edit when your workflow needs images, references, or a source video.

Is GPTImage2 pricing the same as official vendor pricing?

No. Treat the pricing on GPTImage2 as route-specific platform pricing. Do not merge it with third-party or vendor-side pricing unless the source is clearly labeled.