happyhorse
101-skills/superpowers
Generate and edit physically realistic videos with Alibaba HappyHorse 1.0 models via inference.sh CLI.
What is happyhorse?
HappyHorse 1.0 is a suite of Alibaba video generation and editing models accessible through the inference.sh CLI. Use it to create text-to-video, animate images, preserve character consistency across videos, or edit existing videos with natural language instructions at 720P/1080P resolution up to 15 seconds.
- Text-to-video generation with physically realistic motion
- Image-to-video animation from still frames
- Reference-to-video with character preservation from up to 9 reference images
- Natural language video editing with optional character replacement
- Support for multiple aspect ratios (16:9, 9:16, 1:1, 4:3, 3:4, 21:9)
- Reproducible generation via seed parameter and optional watermarking
How to install happyhorse
npx skills add https://github.com/101-skills/superpowers --skill happyhorse- inference.sh CLI (belt) installed and configured
- Valid inference.sh account with login credentials
- Source images or videos in supported formats (JPEG, PNG, WebP for images; MP4/MOV H.264 for videos)
How to use happyhorse
- 1.Install the belt CLI skill: npx skills add belt-sh/cli
- 2.Log in to inference.sh: belt login
- 3.Choose a model based on your task (T2V, I2V, R2V, or Video Edit)
- 4.Prepare your input (text prompt, image URL, video URL, or reference images)
- 5.Run the appropriate belt app command with your parameters
- 6.Monitor progress and retrieve the generated video output
Use cases
- Generate product demo videos from text descriptions
- Create character-consistent social media content using reference images
- Animate still product photos with motion and camera effects
- Edit existing videos to change backgrounds or characters without re-shooting
- Build content pipelines that combine text-to-video with downstream editing
- Content creators and social media managers
- Product marketing and e-commerce teams
- Video editors seeking AI-assisted editing workflows
- Developers building automated video generation pipelines
- Agencies producing character-consistent branded content
happyhorse FAQ
All HappyHorse models support up to 15 seconds of video generation.
Yes, use the R2V (Reference-to-Video) model with up to 9 reference images to maintain character consistency.
720P costs $0.14 per second and 1080P costs $0.24 per second. Video Edit is billed on input + output duration.
Yes, the Video Edit model supports replacing characters by providing reference images along with your editing prompt.
MP4 and MOV files with H.264 encoding are supported for the Video Edit model.
Full instructions (SKILL.md)
Source of truth, from 101-skills/superpowers.
name: happyhorse description: "Generate and edit videos with Alibaba HappyHorse 1.0 models via inference.sh CLI. Models: HappyHorse T2V, I2V, R2V, Video Edit. Capabilities: text-to-video, image-to-video, reference-to-video, video editing with natural language, character preservation, 720P/1080P, up to 15 seconds. Use for: physically realistic video, video editing, character-consistent content, product demos, social media. Triggers: happyhorse, happy horse, alibaba video, happyhorse 1.0, dashscope video, alibaba happyhorse, video editing ai, ai video editor" allowed-tools: Bash(belt *)
Install the belt CLI skill:
npx skills add belt-sh/cli
HappyHorse 1.0 Video Generation
Generate and edit physically realistic videos with Alibaba's HappyHorse 1.0 models via inference.sh CLI.
Quick Start
Requires inference.sh CLI (
belt). Install instructions
belt login
belt app run alibaba/happyhorse-1-0-t2v --input '{"prompt": "a horse galloping across a sunlit meadow"}'
HappyHorse Models
| Model | App ID | Best For |
|---|---|---|
| T2V | alibaba/happyhorse-1-0-t2v | Text-to-video, physically realistic motion |
| I2V | alibaba/happyhorse-1-0-i2v | Animate a single image |
| R2V | alibaba/happyhorse-1-0-r2v | Preserve characters from up to 9 reference images |
| Video Edit | alibaba/happyhorse-1-0-video-edit | Edit existing videos with natural language |
All models support 720P/1080P resolution, up to 15 seconds duration.
Examples
Text-to-Video
belt app run alibaba/happyhorse-1-0-t2v --input '{
"prompt": "a golden retriever running through autumn leaves in a park, slow motion",
"duration": 10,
"resolution": "1080P",
"ratio": "16:9"
}'
Image-to-Video
Animate a still image:
belt app run alibaba/happyhorse-1-0-i2v --input '{
"first_frame": "https://your-image.jpg",
"prompt": "gentle camera zoom, clouds moving in the sky",
"duration": 8,
"resolution": "720P"
}'
Reference-to-Video (Character Preservation)
Generate videos that preserve characters from reference images (up to 9):
belt app run alibaba/happyhorse-1-0-r2v --input '{
"prompt": "a woman walking through a busy market street",
"reference_images": ["https://portrait.jpg"],
"duration": 10,
"resolution": "720P"
}'
Multi-Character Reference
belt app run alibaba/happyhorse-1-0-r2v --input '{
"prompt": "two friends sitting at a cafe having coffee",
"reference_images": ["https://person1.jpg", "https://person2.jpg"],
"ratio": "16:9"
}'
Video Editing
Edit existing videos using natural language instructions:
belt app run alibaba/happyhorse-1-0-video-edit --input '{
"video": "https://your-video.mp4",
"prompt": "change the background to a snowy mountain landscape"
}'
Video Editing with Reference Images
belt app run alibaba/happyhorse-1-0-video-edit --input '{
"video": "https://your-video.mp4",
"prompt": "replace the person with the character from the reference image",
"reference_images": ["https://character.jpg"]
}'
Video Editing with Audio Control
belt app run alibaba/happyhorse-1-0-video-edit --input '{
"video": "https://your-video.mp4",
"prompt": "make the scene look like a rainy day",
"audio_setting": "generate"
}'
Pricing
| Resolution | Price |
|---|---|
| 720P | $0.14 per second |
| 1080P | $0.24 per second |
Video Edit is billed on input + output duration.
Parameters (T2V)
| Parameter | Type | Default | Description |
|---|---|---|---|
prompt | string | required | Text description of the video |
duration | integer | 5 | Duration in seconds (3–15) |
resolution | enum | 720P | 720P or 1080P |
ratio | enum | 16:9 | 16:9, 9:16, 1:1, 4:3, 3:4, 21:9 |
seed | integer | random | Reproducible generation |
watermark | boolean | false | Add HappyHorse watermark |
Parameters (I2V)
| Parameter | Type | Default | Description |
|---|---|---|---|
first_frame | file | required | First frame image (JPEG, PNG, WebP) |
prompt | string | - | Optional text description |
duration | integer | 5 | Duration in seconds (3–15) |
resolution | enum | 720P | 720P or 1080P |
seed | integer | random | Reproducible generation |
Parameters (R2V)
| Parameter | Type | Default | Description |
|---|---|---|---|
prompt | string | required | Text description of the scene |
reference_images | array | required | Up to 9 character reference images |
duration | integer | 5 | Duration in seconds (3–15) |
resolution | enum | 720P | 720P or 1080P |
ratio | enum | 16:9 | 16:9, 9:16, 1:1, 4:3, 3:4, 21:9 |
seed | integer | random | Reproducible generation |
Parameters (Video Edit)
| Parameter | Type | Default | Description |
|---|---|---|---|
video | file | required | Video to edit (MP4/MOV, H.264) |
prompt | string | required | Editing instruction |
reference_images | array | - | Up to 5 reference images |
audio_setting | enum | auto | auto, generate, or keep_original |
resolution | enum | 720P | 720P or 1080P |
seed | integer | random | Reproducible generation |
Search HappyHorse Apps
belt app search "happyhorse"
Related Skills
# Full platform skill (all apps)
npx skills add inference-sh/skills@infsh-cli
# All video generation models
npx skills add inference-sh/skills@ai-video-generation
# Seedance 2.0
npx skills add inference-sh/skills@seedance
# Google Veo
npx skills add inference-sh/skills@google-veo
# Image generation (for image-to-video)
npx skills add inference-sh/skills@ai-image-generation
Browse all video apps: belt app list --category video
Documentation
- Running Apps - How to run apps via CLI
- Streaming Results - Real-time progress updates
- Content Pipeline Example - Building media workflows
Related skills
More from 101-skills/superpowers and the wider catalog.
image-to-video
Convert still images to animated videos with AI models—select the right tool for realistic motion, fabric physics, or versatile animation.
infsh-cli
Run AI apps via inference.sh CLI—image generation, video, LLMs, search, 3D, Twitter automation.
landing-page-design
Design high-converting landing pages with AI-generated visuals and proven layout formulas.
p-image
Generate images fast with Pruna's optimized P-Image models via inference.sh CLI.
p-video
Generate videos from text or images using Pruna's optimized AI models via inference.sh CLI.
p-video-avatar
Generate realistic talking head avatar videos from portrait images—18x faster and 6x cheaper than competitors.