Google adds prompt-based revisions to its Gemini Omni video model

Google has added prompt-based revisions to its Gemini Omni video model, according to a November 7 report from Trend Hunter. In practice, that means a generated clip no longer has to be accepted as-is or thrown away and regenerated from scratch; users can describe the change they want and have the model revise the output.

The outlet frames the update as part of a broader shift toward conversational enterprise video models. Over the past year, competition in video generation has largely centered on image quality and clip length. Once footage can be revised repeatedly through prompts, the competitive ground moves toward workflow: who can best absorb the everyday reality of "not quite right, try again."

For enterprise users, that difference is not cosmetic. Marketing assets, training videos and product walkthroughs usually need several rounds of small adjustments — a reworded line, a different visual emphasis, a tone that matches brand guidelines. If every change requires a full regeneration, the rework cost lands directly on the team using the tool; prompt-based revision is closer to the rhythm of editing a draft.

It also nudges video generation toward the division of labor that enterprise software already assumes. Generation produces the raw material, revision aligns it with brand and review standards, and the instruction history in between leaves a record that teams can work from and audit later.

More editable video also raises the pressure on review and provenance. Buyers tend to ask how revised clips are versioned, how source material is labeled, and whether identifiable watermarks or metadata persist. Those questions are relatively simple for text and considerably harder for video.

Public reporting so far focuses on the capability itself; details on availability, integration paths and enterprise permissioning remain thin. For companies considering video models in a content pipeline, the thing to watch is not only output quality but whether this revision capability fits into existing review and compliance processes.

What comes next is whether Google extends prompt-based revision into a fuller set of collaboration features, and how early users describe the rework it actually saves. The next round of comparison in video generation may hinge less on whether a model can generate and more on how pleasant it is to change what it generated.