How do I use Text to Image?

Create fresh new visuals for your videos with Text to Image! This feature is available Starter, Professional, Enterprise and Agency. Check out below to see how it works! Text to Image is part of our AI credit system. 

• Credits and Model Details
• Steps
• Generate an Image Through Search
• How to Edit Text to Image Generations
• Other Considerations

Check out some helpful tips for your prompt here. 

Credits and Model Details

Steps

Open the Generate Image tab from the asset panel: 

A text box will appear where a prompt can be entered to generate an image or object: 

• Please note, there is a 4000 character limit. 
• Be specific with the style of image, for example 'a realistic orange cat outside in the backyard.'

Click on Generate to create the image or object: 

The generated images or objects will appear in the panel below the prompt box. One image will be generated per prompt. Please allow a few moments for images to be generated: 

Click the three dots on the thumbnail of a generated image an select the option to either Use as Background, Save to Uploads, or Add to collection. 

Generate an Image Through Search

Open the Discover panel, input a search term, and click Enter on the keyboard: 

Select Text to Image at the bottom of the panel: 

The Image Generation panel will open with the searched terms added to the Text to Image prompt box. 

Results will appear underneath the prompt box: 

How to Edit Text to Image Generations

Editing images will cost the same number of credits for each generation. Please see above the Credits and Model Details for more information. This feature only works with the Nano Banana 2 model.

Open the asset panel and navigate to Generate Image or select Text to Image from the Discover section.

Select the Nano Banana 2 model. 

Enter the prompt for the image or upload a reference image to use. 

The image generated will appear below with the changes to the uploaded image.

If you already have a generated image to edit, right click on that image in the studio and select Use in image to image. 

Next, enter the changes to the image and generate. The image will generate with the changes. 

Other Considerations

• There is a 4000 text character limit for the prompt search. 
• Do not ask for text as the model is not set up to provide a text result. 
• Including camera shots in the prompt description may help with the result for backgrounds. For example close up/long shot or top view/front view. 
• Be descriptive of the location, for example 'A cityscape at sunset with cars, front view and people walking on the sidewalk.'
• Colors can be added in the prompt as well. 
• Each prompt is independent of each other so results may vary with each edit that is made. 
• For objects, it's better to search for a stand alone object. 
• NSFW and content that violates Vyond's Acceptable Use Policy is prohibited. NSFW content will not be processed.
• Avoid prompts that reference well-known brands or their logos. Making such references could result in an image that contains intellectual property you don't have the right to use.
• Images are made with generative AI and may not always provide accurate results. 
• The images generated do not necessarily represent view of Vyond.