Alibaba's film-style video model: cinematic camera and pacing, with sound and spoken lines made in the same pass.
Camera moves and pacing lean toward film rather than social video: trailers, drama beats and product reveals.
Every clip has sound made for the scene, and a line in quotes is spoken with the lips moving to it.
16:9, 9:16, 1:1, 4:3 or 3:4, at 5, 8, 10 or 15 seconds. Start from text or from one image.
A man in his thirties in a plain grey t-shirt speaks to camera in a home studio and says clearly: "This is the part everyone gets wrong, and it takes about ten seconds to fix." His mouth matches the words precisely. Soft key light, blurred bookshelf behind, natural presenter delivery, quiet room tone.
Write the line in quotes and HappyHorse says it, with the mouth moving to the words. Turn the sound on to hear it.
A vertical social hook: a woman throws open a fridge door, stares at the empty shelves, turns to camera with a raised eyebrow and says "So. Toast again." Handheld phone-style framing filling the tall frame, kitchen light, comic timing, ambient hum of the fridge.
A fridge opened, an empty shelf, a look to camera and a line, framed for a phone from the first frame.
A trailer shot: a lone armoured figure walks toward camera out of a wall of swirling desert dust as debris lifts around them and a cloak snaps in the wind, the sun a hard white disc behind. Slow motion, low heroic angle, enormous scale, deep bass rumble and wind, no text.
A figure walks out of a wall of desert dust, cloak snapping, with the bass rumble and the wind already on the soundtrack.
A product reveal: a pair of matte black over-ear headphones sits on a dark stone plinth as a ring of light sweeps around them, the ear cups turning slowly, a soft mechanical click as the arm extends. Studio lighting, deep shadows, a low synth tone under the click, no text, no logos.
Headphones on a plinth, a ring of light sweeping round, a soft click as the arm extends. One object, one move, one sound.
Say what happens, how the camera moves and what we hear. Put any spoken line in quotes.
The clip starts on your image, keeping its look and subject.
5, 8, 10 or 15 seconds, in 16:9, 9:16, 1:1, 4:3 or 3:4.
The cost shows on the button before you click. The clip opens in Create and downloads as an MP4.
How a HappyHorse clip starts, and what changed in 1.1, in the maker's own words.
Add a picture and HappyHorse uses it as the first frame of the clip, with an optional prompt for the motion. That is how Alibaba's own API documentation describes the model's image-to-video mode.
“Generate realistic, smooth-motion videos from a first-frame image and an optional text prompt using the HappyHorse model.”HappyHorse makes sound together with the picture. Alibaba lists better audio among the upgrades in version 1.1, the version this page runs.
“HappyHorse 1.1 achieves a comprehensive upgrade over version 1.0 in dynamic expression, role consistency, instruction following, visual texture, and audio capabilities.”HappyHorse is a cinematic video model from Alibaba, launched with a global AI filmmaking competition. HappyHorse 1.1, on this page, makes 5 to 15 second clips with sound from a prompt or a starting image.
No. A 5-second HappyHorse clip is 260 credits, more than a free account or the 200-credit free trial covers. To try cinematic video for less, start with Grok Video, which is free for a 6-second clip.
260 credits for 5 seconds, 415 for 8, 520 for 10 and 780 for 15, sound included. The Generate button shows the cost before you click.
Yes. Clips come with sound made for the scene, and a line you put in quotes is spoken, with the lips moving to it.
Yes. Add an image and the clip starts on it. Image to Video lets you try the same picture on cheaper models first.
5, 8, 10 or 15 seconds in one clip. For a longer piece, make it scene by scene and join the clips in a video editor.
All three come from Chinese labs and make sound with the picture. Wan 2.7 is about half the price of HappyHorse for the same length. Seedance 2.5 runs up to 30 seconds. HappyHorse leans most toward film-style camera and pacing, and handles spoken lines well.
Downloads during the free trial and on paid plans are clean MP4s with no watermark. Downloads on a free account carry a small Vormly watermark.
Yes. Clips you make with HappyHorse here are yours to use in ads, client work, YouTube videos and social posts, with no credit to Vormly needed. New clips are public by default and can appear in Explore with their prompt; on a paid plan you can make them private. A real brand or a famous person's likeness in your prompt is not cleared just because a model made the clip.