Skip to content
Announcements

HappyHorse 1.1 Is Now in Kolbo - Alibaba's Upgraded Video Model with Native Audio

HappyHorse 1.1 by Alibaba is live in Kolbo.AI. Three generation modes including reference-to-video for up to 9 reference images. Native audio, 720p and 1080p, 3 to 15 second clips.

By Zohar - Kolbo.AI Team
NEW MODEL
A meaningful upgrade to Alibaba's top-ranked video model. Better 1080p quality, native audio, and reference-to-video with up to 9 images. Live now.

HappyHorse 1.1, Alibaba's next generation of their widely used video model, is now live in Kolbo.AI. The 1.1 update sharpens the model's output quality at 1080p, reduces the cost of 1080p generation compared to the previous version, and adds reference-to-video as a full generation mode.

Three Ways to Generate

Text to Video. Describe a scene and generate a 3 to 15 second clip at 720p or 1080p, with native audio included. HappyHorse 1.1 handles motion, character consistency, and scene composition well across a wide range of prompt styles.

Image to Video. Start from any still image and animate it into a clip. The model reads the composition and lighting in your image and builds the motion around what is already there.

Reference to Video. Supply up to 9 reference images and HappyHorse 1.1 uses them to build a video that holds together across frames. This mode is especially strong for character consistency and style-matched environments: give it a face, a costume, a setting, and it keeps them coherent throughout the clip.

What Is New in 1.1

Sharper 1080p output. The upgrade improves visual fidelity at 1080p, producing cleaner edges, better texture detail, and more stable motion compared to HappyHorse 1.0.

More affordable 1080p generation. The 1.1 generation costs less per clip at 1080p than the previous version, making the full-resolution mode genuinely practical for regular use.

Native audio. Sound is generated with the video, not added separately. The model produces ambient audio, movement sounds, and environmental atmosphere that matches the visual content.

Flexible durations. Generate clips from 3 to 15 seconds, giving you the right length for everything from a punchy social post to a longer scene.

What It Is Best For

  • Character-consistent video - build scenes with the same character across multiple clips using the reference-to-video mode
  • Social and short-form content - fast, high-quality 1080p clips with built-in audio
  • Scene development - quickly visualize environments, action sequences, and transitions
  • Brand and product video - controlled, visually consistent output at a practical cost

How to Try It

HappyHorse 1.1 is already in your workspace.

  1. Open the Video tool in Kolbo and choose Text to Video
  2. In the model selector, choose HappyHorse 1.1
  3. Write your prompt or switch to Image to Video or Reference to Video mode
  4. Choose your resolution and clip length, then generate

HappyHorse 1.1 is live in your Kolbo workspace right now.

Generate with HappyHorse 1.1 →

Best, Zohar Founder, Kolbo.AI

Tags

videohappyhorsealibabanew-modeltext-to-videoimage-to-videoreference-to-videonative-audio

Related Posts

    We value your privacy

    We use cookies and similar technologies to improve your experience, analyze site traffic, and personalize content. You can choose which types of cookies to accept.