HiDream.ai Launches Global Vivago R1 for End-to-End AI Video Production

HiDream.ai, an artificial intelligence company based in Beijing, has launched its AI video creation platform, vivago R1, on a global scale. The platform aims to streamline the entire video production process, from scriptwriting and storyboarding to video generation and editing, into a single, cohesive workflow.

With vivago R1, creators and businesses can produce finished videos in one production run, eliminating the need to assemble clips from multiple sources. Users can describe their ideas in natural language, provide reference materials, and refine the output through a conversational interface. The platform ensures consistency in characters, settings, and other creative elements throughout the project.

The system employs a series of agents to coordinate the work. A lead agent interprets the brief and develops a production plan, while specialized agents handle tasks such as scripts, storyboards, characters, environments, and audio. HiDream's HD-AgentOS manages the workflow and the context shared across these tasks. Users can start with a written prompt or supply images, video, documents, and other references, which are kept within the project to allow for seamless revision and refinement.

The platform's Chat-First interface is designed to facilitate direct revisions. Users can ask for scene rewrites, storyboard adjustments, or visual element changes without manually configuring a sequence of tools. A library called SkillHub provides reusable workflows for various production tasks, such as developing product ads or narrative shorts.

For those looking to integrate video production into existing workflows, HiDream has made vivago's video capabilities accessible through a command-line interface (CLI). This allows developers and compatible AI agents to incorporate video production into their existing processes. The CLI supports tasks such as initiating video generation, processing media assets, and managing production jobs, making it easier for teams to produce content at scale.

HiDream plans to integrate its native omnimodal video model, HiDream-O1-Video-1.0, into vivago R1 in an upcoming update. This model, which ranked No. 4 globally on the Artificial Analysis Image to Video Leaderboard (With Audio) and No. 7 on the Arena.ai Image-to-Video leaderboard, is designed to improve the handling of multimodal instructions, physical behavior, and audiovisual coherence. The integration will bring these advanced generation capabilities into R1's agent-led production workflow, enhancing the platform's overall performance.

The vivago R1 platform is now available globally at vivago.ai. HiDream has also released an updated version of Gouda for users in China.