Verified, then logged.
Sun 22 Nov 2026 · 00:23 KST / 15:23 UTC한국어 | ENSearch /

Breaking08 / Industry

Google presents VibeStudio video workflow as a developer lab

The lab has developers connect input handling through to Veo rendering; whether a separate commercial app will launch and its own pricing have not been disclosed.

A developer, seen from behind at a desk with a microphone, camera, and photographs, reviews a video timeline and draft poster on a monitor. AI-generated image

Google Codelabs documentation presents VibeStudio as a developer lab for building a video production workflow. Subvolts reported a Google Cloud Tech demonstration of the workflow, dated 2 Nov 2026.

According to Subvolts, the workflow is built with Agent Development Kit (ADK), a software toolkit for building AI agents. Gemini is configured to process voice, photo and video input. Imagen produces posters and Veo renders the video.

The video description also covers waiting for human approval. Deployment to Cloud Run, a managed service for running containerised applications, is listed as well. Annie Wang and Logan Hennessy are named as the presenters.

Cost. The multimodal lab, meaning one that handles more than one type of input, requires a Google Cloud project with billing enabled. The lab document says, "There is one thing this lab cannot hand you: a Google Cloud project with billing on it." Learners have to supply that project themselves.

The multimodal lab separates pauses into waits for a human decision and waits for slow tools. It merges multiple inputs into a single corpus, a body of data collected for search. A separate lab document ends by driving an automated video generation pipeline.

Google Codelabs also published a Korean-language version of the lab document. Unlike the English version, it is labelled Haunted Edition. Whether a separate commercial app will launch and what its own pricing would be have not been disclosed.