Happy Horse 1.1

Happy Horse 1.1 is Alibaba’s talking-video specialist. Speech is generated in-model, with lips that match the words.

Happy Horse 1.1’s signature is native lip-sync. Describe someone speaking and the model generates the voice and the mouth movement together in one pass, without a separate lip-sync step. Ambient sound and effects come with it.

It runs text-to-video and image-to-video, supports 3–15 second shots, and offers a wide aspect ratio set including 21:9 for cinematic framing.

At a glance

  • Type: Text-to-video & image-to-video
  • Duration: 3–15 seconds
  • Resolution: 720p · 1080p
  • Audio: Native synced audio + lip-synced speech
  • Aspect ratios: Incl. 21:9 cinematic (text mode)

What it is best at

  • Lip-synced speech generated in-model, with no separate sync pass
  • Talking-head and dialogue shots from a single prompt
  • Cinematic 21:9 framing available

What it costs

From 77 tokens ($0.77) per clip; 124 tokens ($1.24) at the default settings.

SettingsTokensPrice
3 s · 720p77$0.77
3 s · 1080p98$0.98
5 s · 720p124$1.24
5 s · 1080p159$1.59
8 s · 720p194$1.94
8 s · 1080p248$2.48
10 s · 720p239$2.39
10 s · 1080p307$3.07
15 s · 720p353$3.53
15 s · 1080p454$4.54

Prices come from the live catalog at $0.0100 per token, the pay-as-you-go rate; subscriptions include tokens at a lower rate. The exact price is shown before you generate, and a failed generation is refunded.

Questions

How does Happy Horse 1.1 handle speech?
You write what the character says in the prompt, and the model generates the voice and matching mouth movement together. To sync existing audio to existing footage, use Hiroska’s lip sync model instead.
Can Happy Horse 1.1 animate a photo of a person talking?
Yes. Image-to-video mode animates a still, including making the subject speak, with audio generated natively.
What does Happy Horse 1.1 cost?
It is billed in tokens per generation, scaling with duration and resolution. The picker shows the price for your settings.

Related models