Gemini: Generating AI video with VEO and image references

Gemini Try it →Watch full video →Dean Stokes - Tech Tutorials · Apr 2026

Demo summary

Shows how to use a text prompt and a reference image to generate a custom 8-second video clip of an avatar in a specific setting using VEO.

Step-by-step

Select the VEO option at the top of the screen
Choose an avatar or subject for the video
Write a text prompt describing the action and setting
Click the button under the prompt to add up to three reference images
Click Generate and wait a few minutes for the clip to process
Mouse over the generated clip to preview it
Click Insert and choose to insert into a new scene or the current scene

Options

Prompt for non-avatar subjects like objects (e.g., a treasure chest)
Upload up to three reference images to guide the generation
Insert the clip into a new scene or the current scene

Watch out for

Every clip created by VEO is limited to exactly 8 seconds long

Tips

Use reference images from sources like Unsplash to help the AI generate the specific look you want
If inserting into the current scene, delete the original clip underneath to avoid overlapping media

Highlights

“I'm actually really impressed by that”

All demos from “Complete Guide to Google Vids (2026)”

0:541:16Google Vids interface and timeline overviewThe presenter walks through the Google Vids workspace, explaining the canvas, toolbars, and how to manage scenes and timing on the timeline.Gemini· AI Video Editor
2:161:12Adding and styling text in Google VidsDemonstrates adding a title, selecting Google Fonts like Inter Tight, and using the contextual toolbar to center and bold text.Gemini· AI Video Editor
3:370:56Inserting stock video and reordering layersShows how to search the stock library for London footage, drag it to the canvas, and use the 'send to back' command to place it behind text.Gemini· AI Video Editor
6:082:04Animating objects and text in Google VidsShows how to apply fade-in animations to shapes and text, including separating enter and exit animations and adjusting duration.Gemini· AI Video Editor
8:241:13Generating AI avatars from scriptsDemonstrates selecting a realistic AI avatar (Jayda), entering a script, and generating a video clip with lip-syncing.Gemini· AI Avatar Video Generator
9:451:34Generating AI video with VEO and image referencesCurrentShows how to use a text prompt and a reference image to generate a custom 8-second video clip of an avatar in a specific setting using VEO.Gemini· Image to Video
11:360:28Adding scene transitionsDemonstrates how to preview and apply transitions like 'dissolve' between two contiguous scenes in the timeline.Gemini· AI Video Editor
12:142:42Recording camera and screen with teleprompterWalks through the built-in recorder to capture a Google Maps tab and webcam simultaneously, including the use of the script teleprompter and background blur.Gemini· AI Video Recording
15:040:55Unlinking and adjusting layer timingShows how to use the 'show timing' view to unlink camera and screen recordings to adjust their individual durations and positions.Gemini· AI Video Editor
20:221:42Sharing and exporting Google Vids projectsShows the collaboration sharing settings and the process for downloading an MP4 or exporting directly to YouTube.Gemini· AI Video Editor
Watch “Complete Guide to Google Vids (2026)” →

Image to Video

6:540:26Create a video montage from imagesThe user demonstrates the 'montage' template in Gemini Omni by uploading four images of dogs to generate a cohesive video clip.Paul J Lipsky
2:540:43Animate an image with Google VidsThe user shows how to upload a screenshot and use the 'Animate an image' feature in Google Vids to create a contiguous video scene.Paul J Lipsky
9:451:34Generating AI video with VEO and image referencesCurrentShows how to use a text prompt and a reference image to generate a custom 8-second video clip of an avatar in a specific setting using VEO.Dean Stokes - Tech Tutorials
12:471:34AI image generation and animationThe user generates an AI image of a YouTuber at a beach and then uses the 'convert to video' feature to animate it based on a text prompt.Justin Brown - Primal Video
2:550:30Animate architectural photos with Gemini OmniThe user combines a reference image of a brutalist building with a text prompt to generate a time-lapse video with moving clouds and shadows.AI Master
14:540:37Create video sequences from multiple photosGemini Omni takes five separate photos and generates a coherent 10-second video sequence with smooth transitions and a consistent mood.AI Master
2:192:21Extend AI video duration using image-to-videoThe creator shows a workflow to extend a story by taking a screenshot of the last frame, removing the logo in an external editor, and using the 'animate an image' feature in Google Vids to generate a continuous second clip.Educraft
4:561:19Using 'Ingredients' for object consistency in VeoThe demonstration shows how to upload a specific image (a blue flask) into the 'Ingredients' section of the Veo tool in Google Vids to ensure that specific object appears in the generated video clip.Educraft
6:141:06Animate images with Google Veo 3.1 in Google VidsThe creator demonstrates uploading a still image to Google Vids and using the Veo 3.1 model to generate an animated clip with automatic lip-sync and sound effects.The AI Blueprint
3:001:11Animate an image to continue a sceneThe user takes a screenshot of a generated video frame and uses the 'animate an image' feature to create a follow-up shot where the character turns around.AI Mind Revolution