happyhorse-1.0-t2v is the text-to-video model in HappyHorse 1.0, designed for creative workflows from written concepts to dynamic short films. It focuses on understanding scenes, actions, and visual styles to deliver natural, fluid, detail-rich realistic visuals. On this platform, you can submit generation tasks with prompts, choose duration, resolution, and aspect ratio, making it suitable for creative storyboards, short-film prototypes, and visual direction exploration.
Submit requests to the public API at api.acedata.cloud using the documented parameters, then use the results in your application.
api.acedata.cloudhappyhorse-1.0-t2vInput parameters and result formats vary by service. Use the public API for this model and follow its guide for generation, task retrieval and editing operations.
Text-to-video generation and realistic dynamic performance are capabilities of this model; duration, aspect ratio, and task delivery methods follow the instructions for this platform's invocation endpoint.
The core capability of this model is understanding text descriptions and generating videos. When creating, you can organize the subject, environment, action, and camera movement into a clear sequence—for example, describing a white horse lifting its head in the morning light, its mane swaying in the wind, with the camera slowly pushing in—so the visual focus develops around one consistent intent rather than stacking unrelated requirements.
HappyHorse 1.0's text-to-video capability emphasizes realistic dynamic rendering and natural, fluid, detail-rich results. It is suitable for exploring human or animal movements, environmental changes, and shot atmosphere; prompts can clearly specify motion direction and lighting conditions, then compare visual approaches through different descriptions without first creating a first-frame image.
Generation tasks can be submitted asynchronously, and completion callbacks can also be configured. After saving task_id, an application can query the task or receive the completed result, then read video_url, duration, and resolution. This delivery method is suitable for embedding in an asset workspace, allowing users to continue working on other content after submitting an idea.
Enter the scene, subject action, lighting, and camera movement description for an advertising shot to generate a short video for discussing visual direction. For example, first validate the environmental atmosphere around the product and the shot rhythm, then decide on the formal shooting plan. The deliverable is dynamic proposal material rather than a static storyboard description that relies on text explanations.
Write a prompt around a clear action, select the 9:16 aspect ratio, and generate short video material suited to vertical composition. You can separately try different scenes, lighting, or camera movements, then use selected results for editing. Complete videos involving subtitles, brand marks, or voice-over can be further produced in the post-production workflow.
Organize an individual scene in a story into subject, environment, action, and visual style, then generate a watchable clip to help discuss atmosphere and narrative pacing. It is suitable for early-stage creation when image assets are not yet prepared; when multiple shots are needed, it is recommended to generate and review them segment by segment before combining them into a complete previsualization sequence.
If you already have prompts tailored for happyhorse-1.0-t2v, or want to compare creative options under a fixed model, you can explicitly select this version. happyhorse-1.1-t2v is another text-to-video model in the same series and is also the default generation option; when 1.0 is needed, you should specify model, and must not treat requests with the model omitted as 1.0 calls.
When you only have a text concept and want to freely explore composition, choose t2v for a more direct approach. If you must start from a specified first frame, choose i2v; if you need to use images to constrain characters or style, choose r2v; to modify an existing video, choose video-edit. These are different creation modes, and you cannot simply add image or video fields to t2v as a substitute.
Use prompt to define the subject, action, environment, and camera; this is a text-to-video model and does not automatically switch operations after you submit an image.
Set model=happyhorse-1.0-t2v and action=generate for /happyhorse/videos. Choose an integer duration of 3–15 seconds, 720P or 1080P; text and reference-image generation can set ratio.
Use async or callback_url to connect to background tasks, save task_id and query /happyhorse/tasks; wait for succeeded before reading video_url, and check the people, actions, and audio.
A white horse slowly raises its head in a grassy field, morning light illuminates its mane, the camera gently pushes in, the image is realistic, and a single scene is maintained.
Explicitly select 1.0-t2v and use generate; compare motion and lighting, and do not expect it to automatically switch to I2V after uploading an image.
No. For text-to-video, use action=generate and provide a prompt describing the subject, scene, action, and style. If you want generation to strictly begin from an image, choose the corresponding i2v model instead of adding a first-frame requirement to this text-generation model.
Explicitly specify model=happyhorse-1.0-t2v in the request, together with action=generate. The default text-to-video option is 1.1, so if you need to consistently use the 1.0 workflow, save the complete model configuration rather than saving only the prompt or reusing examples that omit the model.
The platform's text-to-video entry supports 3–15 seconds, with a default of 5 seconds; available aspect ratios are 16:9, 9:16, 1:1, 4:3, and 3:4. It is recommended to determine the aspect ratio based on the final display placement before writing composition descriptions; content exceeding the single-generation range can be split into multiple generations and edited afterward.
You can submit an asynchronous task using async=true, save the task_id, and query it through the task API; you can also provide callback_url to receive the result when the task is complete. Before retrieving the video, check the task status and distinguish between pending, succeeded, and error before arranging downloads and subsequent processing.
It is suitable for generating dynamic visual clips from text, but dialogue, music, or lip-sync should not be regarded as guaranteed delivery capabilities of this model. When producing short films with sound, you can first generate and review the visuals, then add voice-over, music, and sound editing so that audio and visuals each meet the project requirements.
Choose a development approach or application suited to this model and follow its setup guide.
Python SDK Integration Guide
SDK installation, authentication and examplesTypeScript SDK Integration Guide
SDK installation, authentication and examples