Nový Hrádek

[Taylorlly]

Google has introduced Gemini Omni 1.2 Flash, an updated AI video generation model designed to produce longer clips with improved visual consistency and more flexible editing capabilities.

Announced on November 5, 2026, the new model represents another step in Google's efforts to make generative video tools faster, more affordable, and easier to use.

One of the most notable improvements is the ability to generate videos lasting up to 15 seconds. The update also introduces conversational editing, allowing users to refine previously generated videos through additional instructions rather than starting the entire process again.

Longer Video Generation with Greater Flexibility
One of the biggest limitations of many AI video generation systems has been their relatively short output duration.

Gemini Omni 1.2 Flash addresses this limitation by supporting native video generation and editing for clips ranging from 3 to 15 seconds.

This additional duration provides more opportunities for creating short advertisements, product demonstrations, social media content, and visual storytelling.

For example, a creator can generate a product introduction with several visual moments instead of combining multiple extremely short clips.

Users can also specify the desired video duration in one-second increments, giving them more control over the final output.

Conversational Video Editing Changes the Workflow
Another important feature is multi-turn conversational editing.

Traditional AI video workflows often require users to regenerate an entire clip when a small detail is incorrect.

With Gemini Omni 1.2 Flash, users can provide additional instructions to modify an existing generation.

For example, they may request a different camera movement, adjust a visual detail, or refine a specific scene without recreating everything from the beginning.

This approach can reduce unnecessary generation attempts and make the creative process more efficient.

It also allows creators to gradually improve a video through several rounds of feedback.

Better Reference Following and Visual Consistency
Gemini Omni 1.2 Flash also improves its ability to follow reference materials.

Users can guide generation with reference images, short motion clips, and audio cues.

This is particularly useful when creating videos that need to maintain a consistent appearance across multiple scenes.

For example, a business may want to create several promotional clips featuring the same product while preserving its packaging design, colors, and visual identity.

Improved reference adherence can help reduce unwanted visual changes during generation.

However, creators should still review the final video carefully, especially when accurate product details or brand elements are important.

Lower Generation Costs
Google has also reduced the pricing of Gemini Omni 1.2 Flash by approximately 30% compared with Omni 1.1.

The announced 720p generation price is $0.07 per second.

The workflow also supports drafting at lower resolutions before refining and upscaling selected results.

This can be useful for creators who need to test multiple concepts before choosing the best version for final production.

What This Means for AI Video Creators
The release highlights a broader trend in generative video development.

AI video platforms are moving beyond simple text-to-video generation toward more complete creative workflows that include reference control, iterative editing, and resolution improvements.

For users interested in Google's AI ecosystem, 5ggm.com also provides an option to explore Gemini Accounts and related account offers.

Final Thoughts
Gemini Omni 1.2 Flash introduces meaningful improvements to AI video creation through longer clips, conversational editing, stronger reference following, and lower generation costs.

These features could make Google's video generation technology more practical for creators, developers, and businesses producing short-form visual content.