Try Seedance 2.5 Reference-to-Video from ByteDance →
Models
Agents
WorkflowsStudioPricingBlogDocs
ExploreDiscover models by categoryBrowse All ModelsBrowse the complete catalogSee FavoritesSign in to view saved models
OverviewThe platform at a glanceLearnSkills, knowledge, guardrailsAnatomyWhat makes agents reasonBuild Your AgentPick skills, set tier, deploy
Pre-built AgentsBrowse the catalog
Agent Usecases
Ad Campaign ManagerApp Event ManagerApp Review RepliesBarber BookingCustomer Win-BackEcommerce ListingsRestaurant Reviews
Sign InStart Building

Task History

Click to see output list

No tasks yet

Go to Models
Explore models/
Video GenerationActive

alibaba / happyhorse 1.1 reference

happyhorse 1.1 reference

byalibaba

Create 3–15s videos from up to 9 reference images plus a scene description. Designed for character and product consistency at 720p or 1080p.

Image to Video
Model ID
happyhorse 1.1 reference
Provider
alibaba
Updated
1782221348
16
Comments
Average rating : 5 (19 users)
Provideralibaba
Modelhappyhorse 1.1 reference
Image to Video
wiro playground—alibaba/happyhorse 1.1 reference
Reset to defaults
Delete All
0 / 9
Maximum 9 image allowed
Drop image to upload

OR

Click to browse your device

Supports: JPG, JPEG, PNG, GIF, WEBP, HEIC

Required. 1–9 reference images (subjects, objects, scenes). Reference them in the prompt as [Image 1], [Image 2]… by order. Format: JPEG/JPG/PNG/WEBP. Width and height must be at least 300 px. Aspect ratio between 1:2.5 and 2.5:1. Max 20 MB each.

Required. Describes the scene and how the reference images should be used. Up to 5,000 non-Chinese characters or 2,500 Chinese characters.

Sample outputs
Updated 1782221348
## Overview HappyHorse 1.1 Reference is Alibaba’s reference-guided video generator. You provide 1–9 reference images plus a written scene description. The model combines those references into a single short clip with stronger subject consistency. It helps you keep a character, outfit, or product stable across a generated shot. ## What you can build - Character-consistent short scenes for ads, reels, and branded content - Product demo clips that keep packaging shape, labels, and color consistent - Multi-item fashion shots where outfit, accessories, and props stay recognizable - Visual story beats that reuse the same cast across multiple clips - Style-and-subject locked concept pitches for storyboards and pre-vis ## Inputs - 1–9 reference images you want the model to preserve (JPEG/JPG/PNG/WEBP). - Each reference image must be at least 300 px wide and 300 px tall. - Each image must stay within an aspect ratio range of 1:2.5 to 2.5:1. - Each image can be up to 20 MB. - A written scene description that explains the action, camera, lighting, and style. - In the description, point to your references by index, like [Image 1], [Image 2], and so on. - The scene description supports up to 5,000 non‑Chinese characters or 2,500 Chinese characters. - A target clip length between 3 and 15 seconds. - A required output resolution tier: 720p or 1080p. - An optional output aspect ratio. - If you pick auto aspect ratio, the system chooses the closest match from your first reference image. - An optional toggle to add a small “HappyHorse” watermark in the bottom-right corner. - An optional numeric random seed for more repeatable variations. ## Outputs - One generated video clip. - The clip follows your chosen duration and resolution tier. - The frame shape follows your chosen aspect ratio setting. - If enabled, the output includes a visible “HappyHorse” watermark in the bottom-right corner. ## Limitations - Clips are short. The maximum length is 15 seconds. - Output resolution is limited to 720p or 1080p. - Reference fidelity depends on input quality. - Small, blurry, compressed, or low-light reference images can cause identity drift. - Conflicts between your description and the reference images can reduce consistency. - Busy scenes with many subjects raise the chance of warped hands, logos, or fine text. ## Safety & compliance - Platforms commonly apply automated filtering to inputs and generated outputs. - Don’t submit illegal content, sexual content involving minors, or instructions for wrongdoing. - Don’t upload reference images you don’t have rights to use. - Avoid prompts that request real-person impersonation or deceptive political content.

Example prompts

Great starting points for happyhorse 1.1 reference.

The woman from [Image 1] wears the emerald-green silk gown from [Image 2], walking slowly toward the camera in an elegant ballroom. The dress flows with her movement, hair sways gently, soft smile, cinematic warm lighting, slow push-in.Video Generation
The man from [Image 1] leans against the red sports car from [Image 2] on a sunset city street, then turns toward the camera with a slight smirk. Wind moves his jacket, neon reflections shimmer on the car, slow cinematic dolly-in.Video Generation
The cat from [Image 1] curls up on the armchair in the cozy room from [Image 2], slowly blinking and flicking its tail as firelight flickers warmly across its fur. Gentle ambient motion, soft cinematic close-up.Video Generation
The woman from [Image 1] holds the steaming coffee cup from [Image 2] by a rain-streaked café window, lifts it for a slow sip, and smiles softly. Steam curls upward, raindrops trickle down the glass, warm cinematic light, gentle push-in.Video Generation

API quick start

Run happyhorse 1.1 reference with a single API call.

POST https://api.wiro.ai/v1/Run/alibaba/happyhorse 1.1 reference
{
  "prompt": "A woman in a red qipao from [Image 1] unf…",
  "inputImage": "https://your-cdn.com/input.png",
  "resolution": "720P",
  "duration": 5
}
View full API docs

Discover, test, and run AI models, build workflows and agents with one unified API.

All systems operational
WiroAboutBlogCareersContact
ProductModelsAgentsPricingPartnerChangelogStatusFAQ
Getting StartedIntroductionAuthenticationProjectsCode ExamplesWiro MCP ServerSelf-Hosted MCPn8n IntegrationLLMs.txt
API ReferenceModelsRun a ModelModel ParametersTasksLLM & Chat StreamingWebSocketRealtime VoiceFiles
© 2026 Wiro AI. All rights reserved.
PrivacyTermsData Deletion