Text to video
Elo with audio · text-to-video arena
Happy Horse overview
Happy Horse by Alibaba Cloud
Happy Horse by Alibaba Cloud is a next-generation multimodal video generation model designed to bridge the gap between creative intent and production-ready output.
Happy Horse specs
- Model ID
alibaba/happyhorse-1-0- Developer
- Alibaba Cloud
- Input
- Text, Image, Video
- Output
- Video
- Endpoint
POST https://api.aimlapi.com/v2/video/generations
Four ways to generate video
You pick the input that fits your workflow, text, image, reference footage, or an existing clip you want to recut.
Text-to-Video
Describe your scene in natural language and the model handles everything else, motion, camera movement, lighting, ambient sound, and synchronized audio, all generated in a single forward pass. Works at both 720P and 1080P resolution, with durations from 4 to 10 seconds. Particularly strong at physical realism, multi-shot storytelling, and cinematic framing.
Image-to-Video
Supply a single start frame and the model animates outward from it, preserving subject identity and visual style while introducing fluid, physically plausible motion. The optional text prompt guides the direction of movement and camera behavior. This mode achieved an Elo score of 1,416 on Artificial Analysis, the highest recorded for any image-to-video model at the time of launch.
Reference-to-Video
Maintain consistent character identity, style, or scene environment across multiple generated clips. Provide one or more reference images, a character portrait, product shot, or branded environment, and the model uses them as anchors while generating entirely new motion sequences. This is the go-to mode for branded content, character series, and product launch campaigns that need visual consistency without manual retouching.
Video Editing
Transform, recut, or stylistically revise existing video footage through natural language instructions. Tell the model what to change, replace the background, alter the lighting, switch the visual style, or modify specific elements within a scene, and it applies those changes while preserving temporal coherence across frames. Unlike traditional editing pipelines, no timeline experience is required.
Transfusion architecture
Happy Horse uses the Transfusion (Unified Multimodal) framework, a 40-layer self-attention Transformer where text, image, video, and audio tokens sit in a single sequence. The first and last four layers handle modality-specific encoding and decoding; the middle 32 layers share parameters across all modalities.
Use cases
Advertising and marketing: Happy Horse is documented as suited to generating high-quality, fluid, natural videos for advertising and marketing use.
E-commerce showcases: The model is also described as well-suited for e-commerce showcase videos and product-focused creative workflows.
Short-form drama production: Alibaba’s materials say it can be used for short-form drama production, alongside other professional video-creation scenarios.
Social media creative content: The model is documented as suitable for social media creative content and social media marketing.
Happy Horse benchmarks by 
Artificial Analysis
Human-preference results for Happy Horse in text to video, silent video and image to video, next to models available on AI/ML API. Data retrieved Nov 3, 2026.
Models on AI/ML API
Arena results for models you can call with the same API key.
Text to video
Blind votes on clips with generated sound, made from a text prompt. Happy Horse ranks #17 of 31.
Silent video
The same text-to-video task, judged as silent video. Happy Horse ranks #16 of 36.
Image to video
Blind votes on clips with sound animated from a start image. Happy Horse ranks #16 of 29.
Silent image to video
Clips animated from a start image, judged as silent video. Happy Horse ranks #18 of 34.
Happy Horse API pricing
| Usage | Price | Unit |
|---|---|---|
| Video generation | $0.14 | per second |
| Additional | $0.312 | per /sec |
The final price depends on request parameters such as resolution or duration.
Prices from the AI/ML API price list, checked on Nov 4, 2026.
Happy Horse vs other video models: prices
AI/ML API rates for 5 models, per second.
| Model | Per second |
|---|---|
 Happy Horse | $0.14 |
| Wan 3.0 Video | $0.05 |
| Wan 2.7 Image to Video | $0.1 |
| Wan 2.7 Reference to Video | $0.1 |
| Happy Horse 1.1 | $0.14 |
How to use the Happy Horse API
View docs
Build with your agent
Give your agent the model details, code and documentation.
Paste it into your coding agent to build the integration.
1Get your API key
Create API keyexport AIMLAPI_KEY="8f2b4c7d9a1e6b3f…"2Make your first request
curl 'https://api.aimlapi.com/v2/video/generations' \
-H "Authorization: Bearer $AIMLAPI_KEY" \
-H 'Content-Type: application/json' \
-d '{
"model": "alibaba/happyhorse-1-0",
"prompt": "A slow dolly shot through a sunlit greenhouse full of ferns, light dust in the air, natural camera motion"
}'import os
import requests
headers = {
"Authorization": f"Bearer {os.environ['AIMLAPI_KEY']}",
"Content-Type": "application/json",
}
payload = {
"model": "alibaba/happyhorse-1-0",
"prompt": "A slow dolly shot through a sunlit greenhouse full of ferns, light dust in the air, natural camera motion"
}
response = requests.post("https://api.aimlapi.com/v2/video/generations", headers=headers, json=payload, timeout=120)
response.raise_for_status()
print(response.json())const response = await fetch("https://api.aimlapi.com/v2/video/generations", {
method: "POST",
headers: {
Authorization: `Bearer ${process.env.AIMLAPI_KEY}`,
"Content-Type": "application/json",
},
body: JSON.stringify({
"model": "alibaba/happyhorse-1-0",
"prompt": "A slow dolly shot through a sunlit greenhouse full of ferns, light dust in the air, natural camera motion"
}),
});
if (!response.ok) throw new Error(`HTTP ${response.status}`);
console.log(await response.json());The request returns a generation id. Poll GET https://api.aimlapi.com/v2/video/generations?generation_id=<id> until the status is completed.
Models similar to Happy Horse
Related articles
Happy Horse FAQ
What inputs and outputs does Happy Horse support?
Happy Horse takes image, text, video as input and returns video.
How much does Happy Horse cost?
Happy Horse costs $0.14 per second on AI/ML API, billed as you go with no subscription.
How is Happy Horse billed?
Happy Horse costs $0.14 per second on AI/ML API, billed as you go with no subscription.
Can Happy Horse take an existing image as input?
Yes, Happy Horse accepts image input as well as a text prompt.
Who created Happy Horse?
Happy Horse was built by Alibaba Cloud.
How do I call Happy Horse on AI/ML API?
Use alibaba/happyhorse-1-0 as the model id on AI/ML API.
Is Happy Horse available through a single API key?
Yes. Happy Horse is served through AI/ML API, so the same key and endpoint format used for other models applies.
What video generation capabilities does Happy Horse offer?
Happy Horse supports image-to-video, reference-to-video, text-to-video, and video editing.
What is the Happy Horse model ID?
Use alibaba/happyhorse-1-0 as the model parameter. Requests go to https://api.aimlapi.com/v2/video/generations with your AI/ML API key.