HappyHorse 1.0:
Alibaba's #1 AI Video Model

HappyHorse 1.0 is a 15-billion-parameter AI video generation model built by the Future Life Lab inside Alibaba's Taotian Group. It uses a unified single-stream transformer architecture that jointly generates video and synchronized audio from a single prompt, supports text-to-video and image-to-video, produces native 1080p output, and handles lip sync across seven languages. As of April 2026, it ranks #1 on both the text-to-video and image-to-video Artificial Analysis blind-comparison leaderboards.

HappyHorse 1.0 was built by the Future Life Lab inside Alibaba's Taotian Group, the e-commerce arm of Alibaba. The lab is led by Zhang Di, a former Vice President at Kuaishou who previously ran the Kling AI video team. The model launched anonymously on the Artificial Analysis Video Arena before Bloomberg, The Information, CNBC, and Sherwood News confirmed Alibaba as the creator on April 10, 2026.

Yes — HappyHorse 1.0 is published under an Apache-2.0 license and Alibaba has stated the full model weights and GitHub repository will be released. As of April 2026 the practical release is still rolling out: weights and the public repo are marked as coming soon, and most inference providers are still ramping up. The fastest way to actually generate HappyHorse-class video today is to use a hosted platform like Happy Horse AI, which is live and ready to use without a waitlist.

You can try HappyHorse-class AI video generation immediately on Happy Horse AI. The platform brings together leading AI video models in one editor with text-to-video, image-to-video, and full operator controls — aspect ratio, duration, resolution, and audio. New accounts get 50 free credits on signup. No credit card. No waitlist.

On the Artificial Analysis text-to-video leaderboard, HappyHorse 1.0 currently leads with an Elo around 1,388, ahead of OpenAI Sora 2, Google Veo 3.1, ByteDance Seedance 2.0, Kling 3.0 Pro, and PixVerse V6. It also leads the image-to-video leaderboard with an Elo around 1,413. It is the only model currently holding the #1 spot on both tracks simultaneously. These rankings come from blind human votes on output quality, which is significantly harder to game than self-reported benchmarks.

HappyHorse 1.0 outranks ByteDance Seedance 2.0 in both text-to-video and image-to-video on Artificial Analysis. Seedance 2.0 still leads in the audio-enabled image-to-video category by a narrow margin and offers more mature multimodal audio-video integration today. HappyHorse is the leaderboard winner; Seedance is the more battle-tested commercial product. For most creators, the difference is whether you optimize for blind preference quality or for product polish.