Generate visuals with Gen AI Standalone
Gen AI Standalone lets you generate visuals separately from a story, scene, or timeline. Use it when you want to create images, videos, avatars, or visual concepts without attaching them to a specific scene right away. It’s useful for testing ideas, exploring styles, building a visual library, or creating assets you may want to use later in a project.Gen AI Standalone vs. Scene Gen AI
Videfy includes two Gen AI workflows: Gen AI Standalone and Scene Gen AI.Gen AI Standalone
Use Gen AI Standalone when you want to generate visuals independently. Standalone generations are not automatically connected to a scene. This makes it useful for:- Testing creative ideas
- Creating reusable visuals
- Exploring different styles or models
- Generating assets before building a full story
- Creating images or videos outside the scene-by-scene workflow
Scene Gen AI
Use Scene Gen AI when you want to generate visuals for a specific scene in your Storyboard. Scene Gen AI is tied to the selected scene. When you generate or apply a visual from this workflow, it connects to that scene in your project. Use Scene Gen AI when you already know which scene needs a new image or video.Open Gen AI Standalone
You can open Gen AI Standalone from the left-side menu or from inside Videfy Studio.From the left-side menu
- Click the menu icon in the top-left corner.
- Select Gen AI.
From Videfy Studio
You can also access Gen AI Standalone while working inside Videfy Studio.- Open your project in Videfy Studio.
- Click Gen AI in the top navigation.
Choose a generation mode
At the top of the Gen AI Standalone page, select the type of output you want to create. Available modes include:- Text to Image - Generate an image from a written prompt.
- Text to Video - Generate a video from a written prompt.
- Image to Image - Use an existing image as the starting point for a new image.
- Image to Video - Turn an image into a video-style output.
- More - Open additional generation options.
- Audio to Video - Create a video using an audio input.
- Video to Video - Use an existing video as the source for a new video.
- Image to Avatar - Create an avatar-style output from an image.
- Upscale - Increase the quality of an existing visual.
Add your prompt and input
Every generation starts with a prompt. Depending on the mode you choose, you may also need to upload or select an image, video, or audio file as the source input. For better results, include details such as:- Subject
- Style
- Setting
- Mood
- Lighting
- Camera angle
- Colors
- Composition
A modern product photo of a white sneaker on a glass platform, soft studio lighting, clean background, high detailThe more specific your prompt is, the easier it is for the model to generate a result that matches your direction.
Add a negative prompt
Use the Negative Prompt field to describe what you do not want included in the result. For example, you can use it to avoid:- Blurry visuals
- Extra objects
- Unwanted colors
- Distorted details
- Incorrect styles
blurry, low quality, distorted, extra text, dark lightingNegative prompts help refine the output and reduce results that do not match your creative direction.
Adjust generation settings
Before generating, customize the output using the settings below the prompt fields.Style
Choose a visual style for your generation, such as Realistic or another available preset. Styles help guide the overall look and feel of the result.Model
Select the AI model you want to use. Different models may produce different results depending on the prompt, visual type, and style.Provider
The provider shows which AI provider powers the selected model. This may change depending on the model or generation mode you choose.Aspect Ratio
Choose the format for your output. For example:- 16:9 for widescreen or landscape visuals
- 9:16 for vertical content
- 1:1 for square social content
Choose how many images to generate
When using Text to Image or other image-based generation modes, use the output quantity control near the Generate button to choose how many images you want to create. Generating more images gives you more options to compare, but may use more credits. The credit indicator shows the estimated cost before you generate.Generate your output
Once your prompt or input settings are ready:- Review your prompt, input, and settings.
- If available, choose how many images to generate.
- Click Generate.
Organize work with sessions
At the top of the page, you’ll see a session selector, such as Session 1. Sessions help you keep related generations together. This is useful when you’re exploring different ideas, testing prompts, or creating multiple visual directions for the same project. Use sessions to:- Group related generations
- Keep experiments organized
- Return to previous outputs
- Separate work by concept, project, or client
Filter generated content
Below the generation settings, you can filter your results by content type. Available filters may include:- All
- Image
- Video
- Avatar
Use time and advanced filters
You can also narrow your results using additional filters.- Time - Find generations from a specific time range.
- Filter - Apply additional filtering options.
Change the results view
Videfy lets you review generations in different layouts.Grid View
Use Grid View to see generated visuals as thumbnails. This is helpful for quickly scanning and comparing results.Prompt View
Use Prompt View to review generations by the prompts used to create them. This is useful when you want to compare prompt variations or reuse wording from a previous result.Session View
Use Session View to browse generations grouped by session. This is helpful when you organize work by concept, project, or creative direction.Tips for better results
- Write prompts with clear visual details instead of single-word descriptions.
- Use the negative prompt to remove unwanted elements or styles.
- Try different models or styles when the first result does not match your direction.
- Use sessions to keep experiments organized, especially when testing multiple creative concepts.
- Choose the aspect ratio based on where the visual will be used.