InVideo AI adds conversational editing with Gemini Omni Flash
By Steven Van ·
The multi-turn video model lets Invideo Agent One users direct scene, character, and camera changes in plain language.
InVideo AI has switched on Gemini Omni Flash, Google's new conversational video model, inside Invideo Agent One. The company says pairing Gemini Omni with its agent makes editing and creating films, series, ads, and any kind of video feel more intuitive while keeping you firmly in control of the result.
What Gemini Omni Flash brings to InVideo
Gemini Omni Flash is Google's multimodal video model that takes text, images, and video as inputs and generates high-quality clips with native audio. Its headline feature is conversational, multi-turn editing: you can swap characters, relight a scene, change camera angles, or add objects using plain natural language, without re-uploading or re-describing the whole shot. Google positions it as a "Nano Banana for video," grounded in Gemini's real-world knowledge. Inside Invideo Agent One, that editing loop now happens directly in the same interface where you draft and assemble a project.
More intuitive control over full productions
InVideo's pitch has always been turning a single prompt into a finished, edited video, complete with script, footage, music, and transitions. Layering Gemini Omni Flash on top means the refinement stage becomes conversational instead of manual. Rather than dragging clips on a timeline for every tweak, you can direct the agent shot by shot, telling it to change a scene's mood, replace a product, or restyle a sequence, and let it handle the details. That extends across the formats InVideo highlights: films, series, ads, and short-form social video.
Why the timing matters
Google rolled Gemini Omni Flash into public preview via the Gemini API and Google AI Studio, priced around $0.10 per second of output, which the company frames as strong price-performance for generation plus conversational editing. Getting it live inside a mainstream creator tool like InVideo so quickly means everyday marketers, educators, and small teams can try conversational video editing without touching an API or writing code. It lowers the barrier from "AI video is possible" to "AI video is a chat away."
Sources: invideo.io, Google Cloud Blog, The Keyword (blog.google).