alibaba/happyhorse-1.1

Alibaba's Happy Horse 1.1 generates videos from text, animates a single image, or builds a video from multiple reference images. Supports 720p and 1080p, 3-15 second durations, and five aspect ratios.

51 runs
Playground API Examples README

Happy Horse 1.1 is a video generation model from Alibaba. It runs in three modes depending on how many images you give it.

Modes

  • Text-to-video: provide a prompt and no images. Use aspect_ratio to control the shape.
  • Image-to-video: provide a single image. The model animates it as the first frame, and the output adopts the image’s aspect ratio (the aspect_ratio input is ignored).
  • Reference-to-video: provide multiple images (up to 9). The model builds a video that keeps the subjects and scenes from your reference images. Refer to them in the prompt as [Image 1], [Image 2], and so on.

Inputs

  • prompt - text prompt. Required for text-to-video and reference-to-video. Optional when a single image is provided. Up to 2,500 characters.
  • images - input images. None for text-to-video, one to animate (image-to-video), 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 or 1080p. Default 1080p.
  • aspect_ratio - applies to text-to-video and reference-to-video. One of 16:9, 9:16, 1:1, 4:3, 3:4. Default 16:9.
  • duration - integer seconds, 3 through 15. Default 5.
  • seed - integer 0-2,147,483,647. Optional.

Pricing

  • 720p: $0.14 per second of output video
  • 1080p: $0.18 per second of output video
Model created