Soup.video

Happy Horse 1.1

Alibaba's #1-ranked video model. Same form for text-to-video, image-to-video, or up to 9 reference images — with synchronized native audio and multilingual lip-sync.

text-to-video image-to-video reference-to-video
starts from ✦ 20/s
Arrange the skincare serum bottle on the marble surface surrounded by the eucalyptus leaves, white jasmine flowers, and glistening water droplets from the references, forming a fresh premium skincare product scene, with a slow gentle camera move and soft consistent spa lighting.
Medium shot of a professional news anchor at a sleek desk in a modern broadcast studio, cool blue lighting, softly glowing screens behind. 0-5s: He looks into the camera and says in a clear measured voice, "Good evening. Tonight, a breakthrough that could change how millions of us work." 5-10s: He turns slightly toward a second camera, "We'll have the full story, and what it means for you, right after this." Precise lip-sync, subtle studio room tone, crisp broadcast quality, shallow depth of field.
character1 in a cozy dim room strums once, looks up and says in English: "This next one I wrote at three in the morning." Warm practical light, intimate, cinematic, precise lip-sync.

Why Happy Horse 1.1?

Three modes, one form

Text-to-video, image-to-video, or up to 9 reference images.

Native audio

Ambient sound, effects, and dialogue rendered with the picture.

Multilingual lip-sync

English, French, Spanish, Turkish, Japanese, and more.

720p or 1080p

Clips from 3 to 15 seconds, billed per second of output.

Character consistency

Keep subjects and scenes aligned across up to 9 refs.

Pricing

Credits per second
Res Credits
720p ✦ 20/s
1080p ✦ 25/s

How Happy Horse 1.1 works

Happy Horse 1.1 is Alibaba’s top-ranked video model. It runs in three modes depending on how many images you give it — all from the same form, with synchronized native audio and multilingual lip-sync.

Outputs are .mp4, at 720p or 1080p, 3 to 15 seconds. Native audio is generated with the picture.

Capabilities

Use the same prompt bar and settings. Only the reference images change — that is how the model picks a mode.

  • Text-to-video — no images. Write a prompt, pick a ratio, cook.
  • Image-to-video — one still. It becomes the first frame; ratio follows that image.
  • Reference-to-video — two to nine stills. Refer to them in the prompt as [Image 1], [Image 2], and so on.

Text to video

Skip the image button. Describe the scene, camera, and sound in the prompt — including spoken lines for talking-head clips. Set aspect ratio, resolution, and duration, then cook. No input image needed.

  • How — prompt only. No stills attached.
  • Ratio — 16:9, 9:16, 1:1, 4:3, 3:4 (plus 21:9, 9:21, 5:4, 4:5).
  • Native audio — write spoken lines in the prompt for dialogue and lip-sync (English, French, Spanish, Turkish, Japanese, and more).
  • Best for cinematic shot-by-shot sequences and spokesperson / talking-head clips driven purely by text.

Image to video

Attach one PNG, JPEG, BMP, or WebP. That still is the first frame; the prompt describes how it should move and sound. When the image shows a face, multilingual lip-sync matches mouth movement to speech.

  • How — upload or pick one reference, write the motion, cook.
  • Ratio follows the still — the ratio control is ignored.
  • Best for bringing a hero still or keyframe to life — cinematic push-ins, slow orbits, atmospheric scenes, and animating a portrait into a talking shot.

Reference to video

Attach two to nine stills. The model builds a clip that keeps the subjects and scenes from those references. Name them in the prompt in upload order.

  • How — add 2–9 images, then write e.g. [Image 1] dances with [Image 2] under cinematic lighting.
  • Ratio — set it in the form (same options as text-to-video).
  • Best for consistent characters, product scenes, and multi-subject shots that have to match your refs.

Settings

prompt
Required for text-to-video and reference-to-video. Optional for a single image. Up to 2,500 characters.
images
None for text-to-video, one to animate, or up to nine for reference-to-video. JPG, PNG, BMP, or WEBP. Each side at least 300px. Aspect ratio between 1:2.5 and 2.5:1. Up to 10MB each.
resolution
720p (default) or 1080p.
aspect_ratio
Text-to-video and reference-to-video: 16:9, 9:16, 1:1, 4:3, 3:4, 21:9, 9:21, 5:4, 4:5. Ignored for image-to-video — output follows the source image.
duration
Integer seconds, 3 through 15. Default 5.
seed
Optional integer 0–2,147,483,647.