Gemini Omni Flash brings conversational video editing to AI Studio
By Steven Van ·
The new model generates 10-second 720p clips and edits them via prompts, priced at $0.10 per second of output.
Google has launched Gemini Omni Flash, a high-quality, cost-efficient model built for video generation and conversational editing. Announced on June 30, 2026, the model is available today through Google AI Studio and the Gemini API, and it is designed for multimodal workflows where you can refine videos with natural language and simple prompting instead of a traditional timeline editor.
Generate video from text, image, or video inputs
Gemini Omni Flash natively supports high-quality video generation from a combination of text, image, and video inputs. Current generations run to 10 seconds at 720p native resolution, with longer durations and higher resolutions flagged as coming soon. Pricing is set at $0.10 per second of output — the same rate as Veo 3.1 Fast — which puts a 10-second clip at roughly a dollar and makes iterative experimentation affordable inside AI Studio.
Edit videos as a conversation, not a timeline
The headline capability is conversational editing. Rather than dragging clips and keyframes, you describe the change you want and the model applies it while preserving the rest of the footage. Supported edits include background replacement, object removal, style transfer, color grading, element swapping, and time-aware changes to specific segments of a clip. An Interactions API drives this back-and-forth, so you can iteratively refine a result across multiple prompts until it matches your intent.
Built for multimodal creator and developer workflows
Because Omni Flash sits alongside Gemini's other models in AI Studio, creators and developers can prototype in the browser and then move to the Gemini API for production. Google positions the model for image-to-video, reference-to-video, ad creative, social video, product video, and generative media apps — the kinds of fast, high-volume outputs where cost-efficiency and plain-language control matter more than 4K mastering. For teams already building on Gemini, adding video is now a matter of a new model name rather than a new pipeline.
Why it matters for creators
Conversational video editing lowers the skill floor for producing and revising short-form video. Instead of learning a nonlinear editor, a marketer or content creator can generate a base clip and then say what to fix. Combined with AI Studio's free prototyping surface and the pay-per-second API pricing, Gemini Omni Flash makes it realistic to spin up, test, and iterate on video creative directly from a prompt.
Sources: Google blog, Gemini API docs, VentureBeat.