HappyHorse-1.0 Takes First Place, API Coming Soon

HappyHorse-1.0, introduced in early April by Alibaba's ATH division, is an open-source video model that surpassed competitors on the Artificial Analysis leaderboard, notably outperforming ByteDance's Seedance 2.0 in categories without audio. This model, distinguished by its simultaneous audio and video generation capabilities, uses a unified Transformer approach with 15 billion parameters and supports native lip-sync for seven languages. While HappyHorse excels in visual quality, its audiovisual synchronization is on par with Seedance. The model is in private beta, with an API release scheduled for April 30, and it is touted as the first open-source video model of its kind. This launch reflects a broader trend in Chinese AI companies for quiet releases followed by significant announcements, as seen with Xiaomi and Zhipu. The project, led by Zhang Di, highlights Alibaba's dual-engine structure, combining basic research and practical application development. Despite its promising features, challenges such as resource-intensive requirements for local running and fine-tuning remain, with the market showing a positive reaction to its potential impact.

Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.