Gemini Omni 1.2 Flash Adds Native Video Generation and Editing

• Gemini Omni 1.2 Flash is now generally available across the Gemini API and Gemini for business.
• The model generates and edits clips from 3 to 15 seconds, with stronger consistency, reference control, speech and camera movement.
• Google Vids users can create up to three 15-second videos per month at no cost with a Google account or Workspace subscription.

Google has announced Gemini Omni 1.2 Flash, its latest video generation model, with general availability across the Gemini API and Gemini for business. The update adds native video generation and editing, allowing users to create clips between 3 and 15 seconds in one-second steps. Google says the model also improves reference adherence, character and scene consistency, speech generation and camera control.

The release also brings the model to Google Vids. Starting today, anyone with a Google account or a Workspace subscription can generate up to three videos of up to 15 seconds each month at no cost. The update focuses on giving creators more control over a clip and letting them make changes through follow-up prompts instead of starting again.

Gemini Omni 1.2 Flash Adds Native Video Creation and Editing

The main addition in Gemini Omni 1.2 Flash is native video generation and editing. Users can create clips from 3 to 15 seconds, choosing the duration in one-second increments. The model can also work with existing clips, rather than limiting the workflow to generating a new video from a text prompt.

Google says the model is better at keeping characters, objects and scenes consistent from frame to frame. This also applies when a user asks it to create motion between a first and last keyframe or extend an existing clip. These improvements are intended to reduce unwanted changes in a subject’s appearance as the camera or scene moves.

Gemini Omni 1.2 Flash Improves Reference Adherence

Gemini Omni 1.2 Flash can combine character, product and style reference images with motion clips of up to three seconds and audio cues. It is designed to retain important visual details across different camera angles, including faces, clothing, packaging text and brand colours. That can help keep a product or character recognisable throughout a generated sequence.

Google highlights several possible uses for this control. These include virtual try-on clips, product catalogue spins and branded video advertisements. For example, a business could provide product images and a short motion reference, then generate a clip that keeps the packaging and colours aligned with the supplied material. The stronger reference handling is also useful when a video needs to follow a particular visual style.

Multi-Turn Editing Lets Users Refine the Same Clip

Another important change is multi-turn conversational editing. Instead of regenerating a video after every change, users can continue prompting the model to refine the same clip. They can ask it to replace a product or background, change lighting or weather, or adjust the aspect ratio while keeping the rest of the scene consistent.

Google says these follow-up edits work across full 15-second clips. This approach can save time when a video is close to the required result but needs a few targeted changes. It also gives users a more direct way to explore variations without rebuilding the entire sequence after each prompt.

Gemini Omni 1.2 Flash Improves Speech and Camera Control

Speech generation has also been improved, with more fluid dialogue and clearer lip-sync. The model supports multi-speaker conditioning, which can help produce conversations involving more than one character. Google points to multi-character dialogue in Flow, explainers in Gemini for business, and avatars, music and narration in Google Vids as examples of how these capabilities can be used.

Camera control is another focus. The model can create continuous movements such as FPV-style flythroughs, zoom-outs and camera orbits, with fewer abrupt cuts between shots. This gives users more control over how a scene is revealed and can make generated clips feel more connected from one moment to the next.

Google Vids Offers Three Free Videos Each Month

Google Vids now uses Gemini Omni 1.2 Flash for video generation. From 6 November, anyone with a Google account or Workspace subscription can create up to three 15-second videos per month at no cost. The allowance gives individual users and organisations a way to try the updated video tools without paying for each generation.

The model is also available through the Gemini API, including in Google AI Studio, and through Gemini for business in the Google Cloud console. Users can access direct scene creation in Flow. The rollout of 15-second video generation in the Gemini app is starting with Google AI Ultra subscribers.

Pricing and Output Options

Google says 720p video generation is 30% cheaper than with version 1.1, at $0.07 per second. Users can create draft videos at 360p or 720p before deciding whether to make further edits or produce a higher-resolution version. Follow-up editing can also reduce the need to pay for a completely new generation when only a small part of a clip needs to change.

Option	Details
Clip duration	3 to 15 seconds, in one-second increments
Draft resolution	360p and 720p
720p generation price	$0.07 per second, 30% lower than version 1.1
Upscaling	1080p or 4K
Google Vids monthly allowance	Up to three 15-second videos at no cost

After reviewing a draft, users can upscale the result to 1080p or 4K. The combination of lower-cost drafts, targeted follow-up edits and higher-resolution output gives users more choices over how much they spend on a finished clip. The actual cost will depend on the duration and output options selected.

Availability Across Google Products

General availability covers the Gemini API and Gemini for business. Developers can use the API through Google AI Studio, while business customers can access it through the Google Cloud console. Flow supports direct scene creation, and Google Vids has started using the model for its video generation tools. The 15-second video feature in the Gemini app is rolling out to Google AI Ultra subscribers.

Google has also confirmed that generated outputs carry a SynthID watermark. This is part of the company’s approach to identifying AI-generated media. The availability differs by product, so access to the API or Google Vids does not mean that every feature is available in the Gemini app on the same day.

Google has been adding more creation and editing tools to Gemini. Its earlier Gemini file editing update brought document creation and changes into chat, while the Google Connected Apps Platform expanded how services can connect with Gemini and Search. This video update takes the platform further into media creation.

Conclusion

The model brings native video generation and editing, better control over reference images, more consistent scenes and improved speech and camera movement. Its multi-turn editing tools let users refine a clip without regenerating it from scratch, while Google Vids adds a free monthly allowance for short videos. The model is generally available through the Gemini API and Gemini for business, with additional access through Flow, Google Vids and a phased Gemini app rollout.

Frequently Asked Questions