Creative Direction
Visual references for image-led video workflows
Selected frames and stills that help explain how source images, reference material, and finished outputs can shape a Happy Horse video workflow.
Character Reference
Fashion Editorial Frame
A strong character-led still for image-to-video or editorial ad concepts, where styling, posture, and camera language all need to stay readable.
Lifestyle Sequence
Warm Daily-Motion Scene
A softer lifestyle composition that maps well to social content, creator storytelling, and human-centered product demos with more natural motion than a static hero shot.
Available Endpoints
Start building with the Happy Horse 1.1 API
Multiple endpoints for text-to-video, image-to-video, fast preview flows, and async job retrieval. This section is laid out more like a product catalog than raw docs so users can scan what to use first.
Endpoint
Text-to-Video Task
/v1/videos
Submit a prompt-led Happy Horse request by setting model to happyhorse-1.0-t2v. This mode requires prompt and accepts duration, size, aspect_ratio, and seed.
Best for: Use this when the workflow starts from text alone and no source frame or source video is available yet.
Endpoint
Image-to-Video Task
/v1/videos
Submit a first-frame animation request by setting model to happyhorse-1.0-i2v and providing exactly one input image through image, images, first_frame_url, or media.
Best for: Use this when you already have the first frame and need Happy Horse to animate from that fixed starting point.
Endpoint
Reference-to-Video Task
/v1/videos
Submit a reference-image workflow by setting model to happyhorse-1.0-r2v and providing 1 to 9 reference images through reference_image_urls or media.
Best for: Use this when the request needs broader visual guidance from one or more reference images instead of a single locked first frame.
Endpoint
Video Edit Task
/v1/videos
Submit a video edit request by setting model to happyhorse-1.0-video-edit and providing one source video through video_url or media, plus optional reference images.
Best for: Use this when you need to restyle or modify an existing source clip instead of generating the whole clip from scratch.
Endpoint
Task Status
/v1/videos/{task_id}
Poll any submitted Happy Horse task until it reaches a terminal state, then read the hosted result URL from the completed payload.
Best for: Needed for production apps that queue requests, show progress, and persist finished outputs before the hosted URLs expire.
Get started today
Ready to integrate Happy Horse 1.1?
Try the API directly in the console, or reach out to the team for onboarding, pricing, and enterprise setup.
API Documentation
How to get access to Happy Horse 1.1 API
Happy Horse follows the documented ImaRouter video task flow: choose the Happy Horse model id that matches the workflow, submit to /v1/videos, keep the returned id or task_id, then poll /v1/videos/{task_id} until a terminal state is reached.
const apiKey = process.env.IMAROUTER_API_KEY;
async function createHappyHorseClip() {
const createResponse = await fetch("https://api.imarouter.com/v1/videos", {
method: "POST",
headers: {
"Authorization": `Bearer ${apiKey}`,
"Content-Type": "application/json"
},
body: JSON.stringify({
model: "happyhorse-1.0-t2v",
prompt: "Product demo scene with a presenter walking into frame and pointing at the device on a clean studio table",
duration: 6,
size: "1080P",
aspect_ratio: "16:9"
})
});
const task = await createResponse.json();
const taskId = task.task_id ?? task.id;
while (true) {
await new Promise((resolve) => setTimeout(resolve, 3000));
const statusResponse = await fetch(`https://api.imarouter.com/v1/videos/${taskId}`, {
headers: {
"Authorization": `Bearer ${apiKey}`
}
});
const taskState = await statusResponse.json();
const status = taskState.status;
if (status === "failed" || status === "error") {
throw new Error(taskState.error ?? "Happy Horse task failed");
}
if (status === "completed" || status === "succeeded") {
return taskState.video?.url ?? taskState.output?.[0]?.url ?? null;
}
}
}
Async flow
- 1
Choose one of the documented Happy Horse model ids: happyhorse-1.0-t2v, happyhorse-1.0-i2v, happyhorse-1.0-r2v, or happyhorse-1.0-video-edit.
- 2
Submit the request to /v1/videos with the top-level fields required by that mode and store the returned id or task_id.
- 3
Poll /v1/videos/{task_id} every few seconds until status reaches completed, succeeded, failed, or error.
- 4
When the task finishes successfully, persist or download the result URL instead of treating the hosted link as long-term storage.
What Makes It Different
What makes the Happy Horse 1.1 API different
This section is laid out to read more like a product narrative than a feature list. Each row shows a capability, why it matters, and what that looks like in a real workflow.
Character-led motion that feels product-ready
That makes it easier to ship creator, lifestyle, spokesperson, and brand-storytelling experiences that feel closer to production output.
Capability
Character-led motion that feels product-ready
Happy Horse is a good fit when the generated clip needs readable body language, expressive timing, and a stronger sense of human presence than a generic motion demo.
That makes it easier to ship creator, lifestyle, spokesperson, and brand-storytelling experiences that feel closer to production output.
Example scenario
A short-form creator app lets users turn one prompt into a clip that feels polished enough for paid social, organic content, or internal approvals.
Capability
Reference-guided control when the look already matters
Happy Horse supports first-frame animation and broader reference-guided generation, which helps keep styling, framing, and subject identity more stable than prompt-only workflows.
This is especially useful when a brand, creator, or merch team already knows what the scene should look like before motion is added.
Example scenario
A team uploads an approved still or a small set of reference images and generates several motion variants without drifting too far from the original visual direction.
Reference-guided control when the look already matters
This is especially useful when a brand, creator, or merch team already knows what the scene should look like before motion is added.
More than generation: edit existing footage too
That broadens the product surface from simple prompt-to-video demos into a more useful commercial tool for revision, restyling, and creative iteration.
Capability
More than generation: edit existing footage too
Happy Horse is not limited to greenfield generation. It also supports video edit workflows for teams that want to transform footage they already have.
That broadens the product surface from simple prompt-to-video demos into a more useful commercial tool for revision, restyling, and creative iteration.
Example scenario
A creative operations tool starts from an existing clip and uses Happy Horse to clean up the visual treatment, reshape the mood, or align the output with a campaign direction.
Capability
One API path on ImaRouter
ImaRouter gives Happy Horse a production-ready delivery layer, so teams can add it to their app without building their own video queueing and provider glue from scratch.
That makes it easier to launch quickly, compare models later, and keep one integration pattern as your video stack grows.
Example scenario
A platform team adds Happy Horse today, then later introduces more routed video options without throwing away the original integration path.
One API path on ImaRouter
That makes it easier to launch quickly, compare models later, and keep one integration pattern as your video stack grows.
Use Cases
Industries using the Happy Horse 1.1 API
This section keeps the same reusable data model, but the presentation is closer to a grid of industry cards instead of long narrative boxes.
Creator platforms and social video products
Creator and short-form apps
Offer prompt-to-video and image-to-video generation for clips that need stronger character readability and cleaner pacing.
Happy Horse is a sensible choice when users care about presentable motion rather than just seeing any movement at all.
Brand, growth, and paid media teams
Lifestyle and fashion campaigns
Turn approved stills or campaign concepts into motion variants for ads, landing pages, and creative reviews.
Reference-led workflows help keep styling and subject identity more stable across multiple generations.
Ecommerce teams and creative operations
Product storytelling with people
Generate scenes where a person interacts with the product naturally instead of showing only static hero packshots.
This is useful when the final output needs product context and human motion together in one short-form clip.
Product marketers and internal demo teams
AI spokesperson experiments
Prototype short presenter-style clips for internal walkthroughs, campaign concepts, or onboarding content.
Character-led motion is more valuable than abstract cinematic output for teams testing human-centered communication formats.
Agencies and growth studios
Story-first ad generation
Generate multiple short story beats from a single brief, then keep the best direction for further extension or export.
Happy Horse helps when you want a more usable first pass for narrative ad concepts without a custom animation pipeline.
Platform teams and AI product builders
Model-routing video stacks
Expose Happy Horse as one model option inside a broader routed video product that also includes Seedance, Kling, or Wan.
A unified async integration path makes it straightforward to add or swap models based on cost, latency, or creative fit.
Examples
Happy Horse 1.1 API examples
Prompt directions paired with visual reference frames. Use them as inspiration for landing pages, creator tooling, commercial mockups, or API playground defaults.
Text-to-video prompt example
Use when the workflow starts from text only
Pair this with happyhorse-1.0-t2v. It is a plain prompt-only example with no source image or source video required.
Presenter walks into frame, places the product on a clean studio table, looks at camera, subtle push-in, balanced lighting
Image-to-video prompt example
Use when one first frame is already fixed
Pair this with happyhorse-1.0-i2v and provide one first_frame_url, image, images, or media item.
Animate the supplied first frame into a short reveal with gentle camera drift and a clean product-focused finish
Reference-to-video prompt example
Use when multiple references guide the output
Pair this with happyhorse-1.0-r2v and attach 1 to 9 reference_image_urls when a single first frame is too limiting.
Keep the subject identity aligned with the reference images while changing the environment to a brighter studio scene
Video edit prompt example
Use when the task starts from a source clip
Pair this with happyhorse-1.0-video-edit and provide one video_url plus optional reference images.
Preserve the main action, simplify the background, and apply a cleaner studio-grade color treatment
How To Use This API
How to use Happy Horse 1.1 API
This quick-start walkthrough is written to rank for integration-style searches while staying concise enough for busy developers and operators.
- 1
Create your ImaRouter account
Start with an ImaRouter account so you can test Happy Horse and later move the same integration path into production.
- 2
Generate a server-side API key
Create a secure API key for your backend and store it in your deployment environment.
- 3
Choose the Happy Horse mode first
Pick happyhorse-1.0-t2v, happyhorse-1.0-i2v, happyhorse-1.0-r2v, or happyhorse-1.0-video-edit before you build the payload.
- 4
Prepare the required top-level inputs
Match the request to the documented mode: prompt for t2v, one first frame for i2v, 1 to 9 reference images for r2v, or one source video for video edit.
- 5
Submit to /v1/videos and store the task id
Send the payload to the unified video endpoint and keep the returned id or task_id in your own app state.
- 6
Poll terminal status and persist the result
Query /v1/videos/{task_id} until the task reaches completed, succeeded, failed, or error, then save or download the output URL before it expires.
FAQ
Frequently asked questions about Happy Horse 1.1 API
FAQs stay compact and skimmable here. The content is still data-driven for SEO, but the layout is cleaner and less visually heavy.
What is Happy Horse 1.1 API?
Happy Horse 1.1 API on ImaRouter is a public async video task interface exposed through /v1/videos and /v1/videos/{task_id}. It supports text-to-video, image-to-video, reference-image-to-video, and video-edit requests.
Does Happy Horse support image-to-video?
Yes. Use model happyhorse-1.0-i2v for image-to-video and provide exactly one first-frame image through image, images, first_frame_url, or media.
What is the difference between Happy Horse i2v and r2v?
happyhorse-1.0-i2v is for first-frame animation and requires exactly one first-frame image. happyhorse-1.0-r2v is for reference-image-to-video and accepts 1 to 9 reference images plus a required prompt.
Does Happy Horse support video editing?
Yes. Use model happyhorse-1.0-video-edit with a required prompt and one source video via video_url or media. Optional reference images can also be included.
What durations and resolutions are documented?
Happy Horse create requests document durations from 3 to 15 seconds and size values of 720P or 1080P. Video edit does not expose duration in the current schema.
Which aspect ratios are supported?
The current schema documents 16:9, 9:16, 1:1, 4:3, and 3:4 for text-to-video and reference-image-to-video requests.
How does async polling work?
Submit the task to /v1/videos, store id or task_id, then poll /v1/videos/{task_id} every few seconds until the task reaches completed, succeeded, failed, or error.
Should I use metadata or top-level fields?
New clients should prefer the documented top-level fields. Compatibility metadata objects still exist, but top-level payloads are clearer and better aligned with the current public schema.
Do result URLs last forever?
No. ImaRouter's async media result URLs are short-lived, so production systems should download or persist the output instead of treating the hosted URL as permanent storage.
Why use ImaRouter for Happy Horse instead of wiring providers one by one?
ImaRouter combines model routing, five-modality coverage, transparent pricing, automatic failover, and faster new-model onboarding so teams do not have to integrate and monitor providers one by one.
Model Directory
Browse the full model market before you choose your route.
Use the `/models` catalog to scan providers, modalities, reasoning support, context windows, and pricing metadata from a local OpenRouter snapshot. It is the fastest way to compare what exists before you decide which models should be prioritized on ImaRouter.
Get Started
Integrate Happy Horse 1.1 with one documented async video task flow
Use one public endpoint family for Happy Horse creation, reference workflows, and video edit jobs, then expand to other routed models later if needed. Use one API surface for 80+ models across five modalities, with transparent routing, automatic failover, and fast new-model onboarding.