Alibaba Happy Horse V1.1
Happy Horse 1.1 video endpoints for image, reference, and text-driven clips with native audio and lip-sync.
Studio is in early access; availability resolves per project after sign-in.
- Endpoints
- 3
- Eligible
- 3
- Input schemas imported
- 3 / 3
- Delivery
- fal.ai
- Weights
- Not stated
- Developer
- Not stated in source
Overview
Happy Horse 1.1 offers image-to-video, reference-to-video, and text-to-video endpoints. The image endpoint animates stills with synchronized native audio, keeping the source aspect ratio; faces get multilingual lip-sync matched to speech. Aspect ratio derives from the input with no separate control. Prompts guide motion, mood, and sound together.
Capabilities
- Three video endpoints: image, reference, and text inputs
- Native audio rendered with motion
- Aspect-preserving animation with face lip-sync
- Prompt-guided motion, mood, and sound
Best for
- Hero-still and keyframe animation
- Talking-portrait shots
- Atmospheric cinematic moves
Use cases
- Cinematic push-ins and slow orbits
- Portrait-to-talking-shot clips
- Mood-driven scenes with native sound
Endpoints
3 eligible. Cataloged is not the same as executable: Studio resolves which endpoints a project can run.
| Endpoint | Task | Catalog status | Inputs |
|---|---|---|---|
alibaba/happy-horse/v1.1/image-to-video | image to video | Eligible | 6 fields · requires image_url |
alibaba/happy-horse/v1.1/reference-to-video | reference to video | Eligible | 7 fields · requires prompt, image_urls |
alibaba/happy-horse/v1.1/text-to-video | text to video | Eligible | 6 fields · requires prompt |
Questions
Can it change aspect ratio?
No separate control; the ratio comes from the input image.
How is sound handled?
Native audio, ambient sound, and effects render with the video.