VivagoR1 launches as a conversational AI video agent promising deterministic 5-minute outputs
量子位 · wechat · 2026-09-10
HiDream.ai has launched VivagoR1, a conversational video-creation agent in beta for two months, pitched as ending the "gacha" trial-and-error era of AI video. Its predecessor vivago.ai claims 70M+ users globally.
- Positioning: unlike prompt-in/video-out tools or node-based workflow canvases, R1 uses natural-language dialogue as the main entry. A main agent decomposes intent; sub-agents handle script, storyboard, characters, scenes and sound. The chat window is bidirectionally anchored to an interactive canvas with assets auto-synced.
- Capability: the team claims stable 5-minute coherent outputs with character/scene consistency, targeting the L4 stage (single-IP serialized content).
- Hands-on: fed a Three-Body Problem passage, it broke the script down, generated storyboards, image assets and a merged film in one pass — decent quality though with bugs (e.g., abrupt spaceship reveal).
- Cost: roughly 100 RMB per finished video — reasonable for branded content, pricey for individual creators; storyboard editing and cost optimization are planned.
More from Multimodal
- Creator shares full prompt-level workflow for realistic AI videos using GPT Image 2.5 and Seedance 2.5 — socialwithaayan · 2026-09-10
- Seedance 2.5 demo: 'She's an Engineer' video generated by ByteDance's model — Salmaaboukarr · 2026-09-10
- One-Prompt Cardiac Ultrasound Simulation Wows With Surprising Realism — TheMoonMidas · 2026-09-10
- Open-source Dream Loop skill one-shots 3D scenes on a $20 ChatGPT Plus plan, 80% cheaper — tristanbob · 2026-09-10
- Flux 2 Klein image model now runs fully in the browser via WebGPU — radamar · 2026-09-10
- 3D Gaussian Splat Rebuilds Amazon Prime Air Crash Site from New NTSB Footage — bilawalsidhu · 2026-09-10