ทันเอไอ

Google launches Gemini Omni 1.1 Flash for video generation and editing

The new model supports text, images, audio, and video, with controls for start and end points and extending existing scenes

Google officially launched Gemini Omni 1.1 Flash on August 27, 2026, making it generally available as a multimodal model for generative video and video editing.

A key feature is Scene Extension, which can analyze up to 10 seconds of preceding video context before generating a continuation. This helps preserve the atmosphere and visual elements of earlier footage. Users can also control the first and last frames to guide the direction and continuity of transitions.

The model supports video at up to 4K resolution, along with a 360p Draft mode for testing ideas before producing high-quality files. This helps reduce costs while refining outputs and testing prompts. Supported inputs include text, images, audio, and video, allowing users to combine multiple media types when shaping results.

Gemini Omni 1.1 Flash is available through Google AI Studio, Gemini Enterprise Agent Platform, and Google Flow. The launch reflects how AI video tools are evolving from text-to-video generation toward finer control over visual sequences and existing-content editing. However, real-world quality still depends on the footage, prompts, and the user's review and refinement process.