Estimated waiting time 2-4 min
Happy Horse AI Video Generator
Happy Horse 1.0 by Alibaba's Future Life Laboratory — the #1 model on the Artificial Analysis Video Arena. Turn text or images into physically realistic 1080p video that beats Seedance 2.0 and Kling 3.0 in blind tests.
Frequently asked questions
What is Happy Horse 1.0?
Happy Horse 1.0 is Alibaba's latest video generation model, built by the Future Life Laboratory inside the Taotian Group and led by Zhang Di — the architect behind Kling AI. It generates physically realistic 1080p video from a text prompt or a single image and currently holds the #1 spot on the Artificial Analysis Video Arena.
Why does it top the leaderboard?
On the Artificial Analysis Video Arena (a blind, Elo-based human voting system), Happy Horse 1.0 leads Seedance 2.0, SkyReels V4, and Kling 3.0 Pro by the largest margin recorded in the arena's history — both for text-to-video and image-to-video. Real users prefer its output roughly 60-65% of the time in head-to-head comparisons.
What modes are supported on Viw AI?
Two: Text-to-Video and Image-to-Video. Both run through Happy Horse's unified single-stream pipeline, so you get the same quality whether you start from a prompt or a first frame. Image-to-Video derives the aspect ratio automatically from the image you upload.
What durations and resolutions can I pick?
Anywhere from 3 to 15 seconds, set with a slider, at 720p or 1080p. Pricing is per second: 720p costs 28 credits per second, 1080p costs 48 credits per second — so a 5s 720p clip is 140 credits, and a 10s 1080p clip is 480 credits.
How is Happy Horse different from Wan 2.6?
Both come from Alibaba, but Happy Horse uses a newer 15B-parameter single-stream Transformer with 8-step DMD-2 distilled inference — roughly 38 seconds for a 5-second 1080p clip on a single H100. It generates faster than the multi-stage Wan family and leads it on independent blind benchmarks.
Will my video have a watermark?
No. Videos generated on Viw AI export without a Happy Horse watermark, so you can use them in your own projects directly.
What aspect ratios does Text-to-Video support?
16:9, 9:16, 1:1, 4:3, and 3:4. Pick the one that fits your target platform — 9:16 for TikTok, Reels, and Shorts; 16:9 for YouTube and product shots; 1:1 for Instagram feed posts.
How do I write a good prompt?
Describe the subject, the action, the setting, and the camera together. Specific cues like "slow dolly-in," "soft side lighting," or "rain-slick pavement" produce more consistent results than vague descriptions. Happy Horse accepts prompts in any language, so you can write in your native one.
