Happy Horse 1.1: The Essentials

Last updated: June 22, 2026

asset_YMLSGcyBnCdSQnA3W4bWy4ea_A clean, minimalist conceptual banner design, viewed from an overhead perspective on a modern, subtly textured desk. Several sleek digital screens are arranged, showcasing video gener.png

Happy Horse 1.1 by Alibaba generates short cinematic clips with synchronized native audio and multilingual lip-sync in one pass. Describe the scene and the sound together, animate a still, or lock in up to nine reference characters with the R2V model. Clips run 3 to 15 seconds at 720p or 1080p.

Video generated using Happy Horse 1.1

Which Model Should I Use?

Model

ID

Input

Best for

Happy Horse 1.1 T2V / I2V

model_alibaba-happy-horse-1.1

Text prompt, optional first-frame image

Scenes, action, talking heads, and animating a still without character refs

Happy Horse 1.1 R2V R2V

model_alibaba-happy-horse-reference-to-video-1.1

1 to 9 reference images + prompt with character1…character9

Consistent characters across a clip, multi-subject dialogue, brand mascots

Rule of thumb: start with Happy Horse 1.1 when you want a scene from words or a single still. Switch to R2V when specific faces or characters must stay recognizable and you can supply portrait references.

How to Use the Model

Text-to-Video with Native Audio

Write one prompt that covers visuals, camera, dialogue, and sound. Mention lip-sync when a subject speaks to camera. Clips from 11 to 15 seconds give talking heads room to breathe.

prompt: A professional news anchor at a studio desk speaks directly to camera with clear lip-sync, saying welcome to tonight's briefing, subtle camera push-in, broadcast lighting, confident delivery with native studio ambience
duration: 12
resolution: 1080P
aspectRatio: 16:9

Broadcast lip-sync from text only.

prompt: A punk rock singer screams into a microphone on a packed stage, strobe lights, crowd surfing energy, raw live performance with native audio and crowd roar
duration: 14
resolution: 1080P
aspectRatio: 16:9

Live performance with native crowd audio.

Image-to-Video from a First Frame

Upload or generate a still, pass it as image, and describe how the scene should move and sound. Omit image for pure text-to-video.

image: asset_PAeSfqoohnPELrqZvh9YWXBs
prompt: The fantasy knight raises a sword triumphantly, cape billowing in wind, epic orchestral energy, slow dramatic camera orbit, particles and light flares, game cinematic trailer mood
duration: 13
resolution: 1080P
aspectRatio: 9:16

Scene animated from a Gemini 3.1 still.

image: asset_beqrRp91ddfYj7rCv1siupa2
prompt: The surfer rides a towering wave, carving through spray as the camera tracks alongside, ocean roar and wind native audio, epic sports cinematic
duration: 14
resolution: 1080P
aspectRatio: 16:9

Sports action from a wave still. asset_4QuuyxRD3WQP7QC8eNXnXRog

Reference-to-Video with character1 Labels

On the R2V model, add 1 to 9 portrait references. Refer to them as character1, character2, and so on in the prompt. The number matches the upload order: first image is character1, second is character2.

Reference order matters. If you swap the images, swap the character numbers in the prompt. Each reference set should be unique per generation when you need distinct showcase examples.

referenceImages: [asset_EKQofgsyhB6QQ6JPrzGAutvY]
prompt: character1 floats inside a space station module and delivers a calm mission briefing to camera, Earth visible through the window, subtle zero-gravity movement, native audio with lip-sync
duration: 11
resolution: 1080P
aspectRatio: 16:9

Single-character briefing with lip-sync. asset_7ogTYb8XLY9BerqkYDkDMHxF

referenceImages: [asset_cRVp3B98JPsUYL18V8TeJpzV, asset_LKPMzeg32Qs6NAqxqmGu4gUW]
prompt: character1 interrogates character2 across a desk in a smoky noir office, venetian blind shadows, tense lip-sync dialogue, rain on window native audio
duration: 14
resolution: 1080P
aspectRatio: 16:9

Two-character dialogue scene. asset_4sxJKXNqJ1GvUxAoRari4pKQ

Chef + cyclist duo · asset_nFUNtZoUfDojrKjyeSEymGXG

Three-character royal ball · asset_HFXGPfkr6YwoqaUHU92gDzmb

Building Reference Portraits

For R2V, generate clean character headshots with GPT Image 2 or Gemini 3.1 Flash, then use the resulting assets as referenceImages. Square 1:1 portraits at 1024 px or above work well. Use a fresh reference set for each example when you need distinct characters.

image.png

Parameters

Both models share the same output controls. R2V replaces the optional image field with required referenceImages.

prompt

Required. Up to 2500 characters. Describe scene, motion, camera, dialogue, and native audio together. On R2V, label subjects as character1, character2, and so on to match reference order. See the news-anchor and noir examples above.

image

Optional on Happy Horse 1.1 only. A first-frame still to animate forward. Leave empty for text-to-video. When set, output aspect ratio may follow the image shape. See the Tokyo neon and surfer wave examples.

referenceImages

Required on R2V. An array of 1 to 9 portrait asset IDs. Order sets character numbers in the prompt. Each image should be at least 400 px on the short side. See the astronaut solo and three-character ball examples.

duration

Optional. Clip length in seconds, 3 to 15. Default is 5. For talking heads and dialogue, 11 to 15 seconds tested best in onboarding. See the 12 s anchor and 15 s royal-ball clips.

resolution

Optional. 720P or 1080P. Default is 1080P. All examples in this article used 1080P.

aspectRatio

Optional. Nine presets: 16:9, 4:3, 1:1, 9:16, 3:4, 4:5, 5:4, 9:21, 21:9. Default is 16:9. Match social placement: 9:16 for Stories, 21:9 for ultra-wide trailers.

seed

Optional. Integer from 0 to 2147483647. Reuse the same seed with identical settings to reproduce a clip. Leave empty for a new result each run.

Use Cases

  • Social and ads: Vertical or widescreen product and lifestyle clips with voiceover and ambient sound baked in.

  • Games: Short cinematics, NPC dialogue beats, and trailer moments with synchronized speech.

  • Marketing: Talking-head explainers, event promos, and character-led campaigns via R2V.

  • Film and previs: Blocking dialogue scenes and atmosphere tests before full production.

  • Education: Instructor-style clips and multilingual lip-sync demos from a single prompt.

  • Content creation: Animate stills from GPT Image 2 or Gemini 3.1 into motion clips with sound.

Tips for Better Results

  1. Pair visuals and audio in one prompt. Name dialogue, ambience, and music cues explicitly; the model generates them together.

  2. Use 11 to 15 seconds for speech. Shorter clips cut off mid-sentence on talking-head scenes.

  3. Mention lip-sync for speakers. Phrases like "clear lip-sync" or "delivers a briefing to camera" improve mouth movement.

  4. Keep R2V references as clean portraits. Generate square headshots with GPT Image 2 or Gemini 3.1 Flash before running R2V.

  5. Match character numbers to image order. The first reference is always character1; swap both images and labels together.

  6. Start at 1080P for finals. Drop to 720P only when iterating quickly on composition.

  7. Omit image for pure T2V. On Happy Horse 1.1, leave the image field empty when you do not need a locked first frame.

Known Limitations

  • Maximum clip length is 15 seconds. Plan multi-beat stories as separate clips or extend in an editor.

  • R2V requires character labels. Plain names in the prompt without character1 syntax will not bind to reference images.

  • First-frame aspect ratio override the setting. On I2V, the output follow the uploaded still's shape.

  • Platform auto-captions are not reliable for QA. Review lip-sync and audio by watching the clip, not the auto-generated caption text.

Open the models: Happy Horse 1.1 · Happy Horse 1.1 R2V