Google Gemini Omni 1.2 Flash Extends AI Video Clips to 15 Seconds

Longer Clips, Smarter Edits, Fewer Glitches

Google has released Gemini Omni 1.2 Flash, a new AI video model that generates clips up to 15 seconds long. The update launched on Thursday and introduces conversational editing features. Creators can now direct scene changes through natural language commands. This addresses a common frustration where AI videos cut off before the action peaks.

The previous version, Omni 1.1, had a shorter maximum duration. The new model doubles that limit, giving storytellers more room to develop scenes. Users can now ask the AI to „make the car drive fasteror ”add rain to the backgroundthrough simple text prompts.

What Does This Mean for Content Creators?

The upgrade focuses on three core improvements: extended runtime, conversational editing, and reduced generation errors. Shorter clips forced creators to stitch multiple generations together, creating visible seams. The 15-second cap eliminates that patchwork approach. The conversational editor lets users refine output iteratively without regenerating the entire clip.

For social media producers, the longer clips align with platform trends. TikTok and Instagram Reels favor 15-second formats. Creators can now generate complete narrative beats in a single pass. The conversational editing feature lowers the technical barrier for non-professionals.

How long are the new AI video clips? Gemini Omni 1.2 Flash generates videos up to 15 seconds long, double the previous limit.

Frequently Asked Questions

Can users edit the generated videos conversationally? Yes. The model supports natural language editing commands to adjust scenes, pacing, and visual details without regenerating the full clip.

When was this update released? Google rolled out Gemini Omni 1.2 Flash on Thursday.