Create AI video, images, music, and voice
Lab gives you direct access to AI generation tools in one workspace. Choose a model for your task, write a prompt, and upload images, video, or audio as references where supported. Adjust the available settings, preview your results, and download the media you create.
Explore the tools before signing in. Sign in to start a generation and revisit your run history. You can switch between video, image, music, and voice tools as you build your next creative.
V4V Lab
Gemini Omni Flash 1.1
Generate up to 4K video from first/last frames or multimodal image, video, audio ID, and character ID references.
First Frame
Optional starting frame. First-frame mode cannot be combined with image, video, audio ID, or character ID references.
Last Frame
Optional ending frame. A first frame is required whenever a last frame is provided.
Reference Images
Optional reference images. Gemini visual quota is images + video*2 + characters <= 7. Cannot be used in first-frame mode.
0/7
Drop media here
Reference Video
Optional source video reference. Max 1. Cannot be used in first-frame mode.
0/1
Drop media here
Reference Audio
Optional Gemini Omni audio IDs. Max 3. Cannot be used in first-frame mode.
0/3
Character IDs
Optional Gemini Omni character IDs. Max 3. Cannot be used in first-frame mode.
Create a character ID from an avatar image
0/3
Duration
Output video duration in seconds. Allowed values: 4, 6, 8, or 10. Ignored when a reference video is provided.
Aspect Ratio
Output aspect ratio. Allowed values: 16:9 or 9:16.
Resolution
Target output resolution, from fast 360p drafts through 4K.
Advanced
Less frequently used model settings.
84.00 credits / run