Gemini Omni Gets Video Editing - Edit Any Video With a Typed Prompt

Google's Gemini Omni model family has expanded in Kolbo. Alongside the original multimodal video generation that accepts text, image, video, and voice as inputs, Gemini Omni now includes a dedicated video editing mode and a full image-to-video mode.

Gemini Omni Edit: Change Anything in a Video With Text

The most significant addition is Gemini Omni Edit: upload any video, type a description of what you want changed, and the model applies the edit and returns the result. Native audio is preserved. Output is 720p.

What you can do with it:

• Swap a wardrobe mid-scene: change a character's outfit to any style or color without touching the timeline
• Restyle a scene: shift the visual aesthetic, season, lighting mood, or environment
• Modify specific elements: change colors, materials, or props in the shot
• Recompose the visual feel: apply a different tone or era to existing footage

The edit does not require masking, selections, or frame-by-frame work. You describe the change in plain language and the model handles it. A single generation costs a flat credit amount.

Note: Gemini Omni Edit is geo-restricted in some regions. If it is unavailable from your location, the tool will indicate that.

Gemini Omni Image-to-Video: Animate a Still Image

The new image-to-video mode takes a single still image and generates a video clip from it. Upload your image, add a motion prompt or description, and Gemini Omni creates a 3 to 10 second clip at 720p. Supports 16:9 and 9:16 aspect ratios.

This is separate from the Elements mode and is available directly in the Image-to-Video tool.

The Original Gemini Omni: Multi-Input Video Generation

The original Gemini Omni Flash mode in Kolbo's Elements tool remains unchanged. It accepts any combination of text, images, video clips, and voice input and generates video that integrates all of them coherently. It is still the most flexible input-to-video model in Kolbo for mixed-media projects.

How to Try Each Mode

Gemini Omni Edit:

1. Open Video Tools and choose Video-to-Video
2. Select Gemini Omni Edit from the model list
3. Upload your source video and type the change you want
4. Generate

Gemini Omni Image-to-Video:

1. Open Video Tools and choose Image to Video
2. Select Gemini Omni Video from the model list
3. Upload a still image and add your motion prompt
4. Set duration (3-10 seconds) and generate

Original Gemini Omni (multimodal):

1. Open Video Tools and choose Elements
2. Select Gemini Omni in the model selector
3. Add any combination of inputs and generate

Best, Zohar Founder, Kolbo.AI