- Gemini Omni 1.2 Flash generates videos up to 15 seconds long, compared with the previous 10 second limit.
- Google has reduced pricing by 30%, bringing 720p video generation to $0.07 per second.
- Improved reference following helps maintain consistent characters, products, clothing and visual details.
- Conversational editing allows users to refine existing clips, while the model expands across Google Cloud, Gemini and creative applications.
One of the biggest improvements in Gemini Omni 1.2 Flash is its ability to generate videos lasting up to 15 seconds in a single request.
Although five additional seconds might not sound particularly significant, the improvement could make a noticeable difference for creators. Longer clips offer more room for storytelling, product demonstrations, advertising campaigns and educational content without requiring multiple generations to complete a sequence.
Users can specify video durations in one second increments, starting at three seconds and extending to 15 seconds.
Google has also made the model more affordable. Gemini Omni 1.2 Flash costs $0.07 per second for 720p video, representing a 30% reduction compared with the previous generation.
The company is encouraging creators to adopt a more efficient production process. Instead of immediately generating high resolution videos, users can begin with lower resolution drafts at 360p or 720p.
Once the composition and movement are satisfactory, they can make targeted edits before upscaling the finished result to 1080p or 4K.
This approach could help businesses reduce the cost of experimenting with different video concepts while avoiding unnecessary generations.
Better reference following and improved visual consistency
AI video generation has made considerable progress, but maintaining consistent characters, objects and visual details across longer clips remains challenging.
Gemini Omni 1.2 Flash is designed to address this problem.
The model can work with character references, product photographs, style images, short motion clips and audio cues. Google says it can preserve important visual details throughout a generated sequence, including facial features, clothing, product packaging and brand colours.
This could prove particularly useful for businesses creating promotional videos.
For example, a company could provide product photographs and reference images of a model before generating a virtual try on sequence. The model is designed to preserve the clothing’s appearance, including its cut, fabric movement and details from different angles.
That could make AI video generation more practical for online retailers, fashion companies and advertising agencies.
Google also highlights improvements in maintaining consistency between the beginning and end of a clip.
The model can connect different keyframes, extend existing footage and produce smoother transitions while reducing unwanted visual changes.
This is important because AI generated videos can sometimes suffer from visual drift, where characters or objects gradually change appearance as the scene progresses.
Gemini Omni 1.2 Flash aims to make those transitions more reliable, helping creators produce videos that look closer to professionally directed footage.
Conversational editing makes video production more flexible
Another major addition is multi turn conversational editing.
Rather than starting over whenever a video needs a change, users can continue refining the same clip through follow up instructions.
For instance, someone creating an advertisement could ask the model to replace the background, adjust the lighting, change the weather or modify the aspect ratio.
The model is designed to preserve the original characters, performances and camera framing while applying the requested changes.
This could be particularly useful for advertising teams that need multiple versions of the same campaign.
A single video could be adapted for different audiences, social media platforms or marketing placements without requiring a completely new generation each time.
Google says conversational editing works across full 15 second clips, giving users more flexibility when refining their projects.
The model also introduces improvements to dialogue generation, including more natural speech, clearer lip syncing and better handling of multiple speakers.
These changes could benefit video explainers, animated scenes and avatar based presentations.
Google is also working to improve camera control. Gemini Omni 1.2 Flash can reportedly handle more complex movements, including cinematic zooms, sweeping camera flights and continuous transitions.
The goal is to reduce unwanted jump cuts and maintain a more natural sense of movement throughout a scene.
Gemini Omni 1.2 Flash expands across Google’s creative ecosystem
Google is positioning Gemini Omni 1.2 Flash as a tool for both individual creators and professional production teams.
The model is available through the Gemini API and Google Cloud, with enterprise infrastructure and security supporting business deployments.
Google also says generated videos include an imperceptible SynthID watermark, helping identify content produced using its AI technology.
The model is expanding across several Google products.
Developers can experiment with it through Google AI Studio and integrate its capabilities into applications using the Gemini API.
Businesses can access it through Google Cloud, while creative professionals can use it through supported applications such as Adobe Firefly and Figma Weave.
Google Vids is also receiving access to 15 second video generation, with eligible users able to create clips without additional charges under supported account arrangements.
The Gemini app is expected to receive the new video generation capabilities as the rollout continues, with access initially expanding to Google AI Ultra subscribers.
Google says the model is particularly suitable for businesses, developers and creators producing videos at scale.
However, the company acknowledges that some areas still need improvement. Fast sports action, rapid script changes and complex physical interactions remain challenging for AI video generation.
Despite those limitations, Gemini Omni 1.2 Flash represents a meaningful upgrade for creators looking for better control, longer clips and lower production costs.
Its combination of conversational editing, stronger reference following and more affordable generation could make AI video production more practical for everyday creative work.
Follow TechBSB For More Updates
