Google Vids adds Gemini Omni generation and personal avatars for paid users

The new tools let eligible subscribers create, revise and present videos from prompts, image references, a selfie and a voice sample.

Google is rolling out Gemini Omni and personal avatars in Google Vids, extending the Workspace video tool from manual assembly and individual clip generation toward prompt-based creation, editing and presenter-style output.

Gemini Omni lets users generate videos by describing the desired result in natural language. Users can also add image references, including photographs or rough sketches, to guide the generated video.

The update also adds conversational editing. Google says users can ask Vids to change a background, correct lighting or add effects to an existing clip, whether it was generated in Vids or recorded on a phone. Omni supports iterative edits, allowing users to revise a video step by step rather than regenerate the entire asset after each change.

Google previously rolled out Veo 3.1 video generation to Vids users in February. Omni expands the workflow beyond generating an initial clip, with text-and-image input and natural-language revision controls integrated into the Vids editor.

The second feature, personal avatars, creates a digital presenter from a user-provided selfie and short voice recording. After setup, a user can type a script and have the avatar deliver it without recording a new on-camera video.

For teams producing routine updates, training snippets or internal announcements, the practical appeal is reducing the operational friction around video production. A product manager or subject-matter expert can prepare a scripted message without arranging a shoot, while editors can use prompt-driven changes for common revisions such as visual cleanup or background replacement.

That does not remove the need for review. Prompt-based generation can shorten iteration cycles, but organizations still need to check whether generated visuals, edits and avatar delivery match brand requirements and accurately represent the intended message. Google did not publish benchmark scores, model parameter counts, generation times or pricing changes for Gemini Omni in Vids.

Gemini Omni and personal avatars are available in Google Vids for Google AI Pro and Ultra subscribers and Google Workspace business customers, Google said. The company also said personal-avatar access is currently limited to a subset of users, but the published announcement does not fully specify that eligibility restriction.

Rollout timing by region, the complete set of supported Workspace plans and other limitations on personal avatars are therefore still unclear. Google’s announcement confirms the tools are being rolled out, but not the precise availability schedule or all deployment conditions.