AI Visualizer: Generate tab

The Generate tab provides controls for defining the AI image, using a combination of text prompts, source images, and additional settings such as aspect ratio.

Parameter	Description
Prompt	Enter a description of the image to generate. The prompt should use natural language, but be precise and detailed for the best results. It can be lengthy, such as a paragraph. The description should refer to any image prompts provided, but the AI model does not understand references to them by number. Rather than "apply the pattern in the second image prompt to the sofa in the first image prompt," for example, you should say something like "apply the pattern to the sofa." The visualizer model should be able to identify which image provides a pattern and which has a sofa to use in generating the image. See: Tips for writing a good prompt.
Negative Prompt	Optionally, enter a description in natural language of anything that should be omitted from the image. Sometimes AI adds a generic logo placeholder to the image; putting "logo" in this field may help eliminate that. If not, try editing the image to remove the logo.
Image Prompts	You can optionally use multiple images from Archicad or an imported image file to provide context for the image generation; the source images display here. These image prompts can guide the composition of the generated image, so you can use a design created in Archicad as the basis for the image. They can also provide style elements such as specific textures or colors or a mood board, provide image props to be added to an image, and so on. The AI Visualizer works by detecting edges, colors, and textures if image prompts are used, to infer composition and materials. For example, an area of green near the bottom of the image is interpreted as grass. The first image prompt, when used, also sets the aspect ratio of the generated image. To import an image, click the upload button and select the desired image file. To use the current Archicad view: set the desired view and click the view snapshot button. The image prompts are used in combination with text prompts and the other settings to influence how the image is generated. A text prompt of at least one word is required. To clear the image, click the X. When no image prompt is used, an image is generated based on the text prompt and other settings.
Enable seamless tiling	Generates a seamless tiled pattern for use as wallpaper, image fills, textures, and so on. If this is selected, the text prompt should be a string of descriptive text, with a comma between each segment of the prompt, rather than a paragraph using natural language.
Creativity (applies to tiling)	Indicates how far the generated tiled pattern should deviate from the source image, or how conceptual a tile based only on a text prompt should be. Either click on the percentage number and enter a value, or drag to set the desired percentage. Higher values should produce more colorful and creative results.
Aspect Ratio	When no image prompts are loaded, select the aspect ratio of the generated image. When one or more image prompts are loaded: the aspect ratio automatically matches the first image prompt.
Generate	Generates the image. Generating a new image discards the previous image; if you want to save an image for future reference, use one of the options on the Save tab before generating a new image.
Previous image	Click to see the last image generated before your most recent edit. Drag the arrow over the image box to show parts of both current and previous image.
Current image	Click to see the most recently generated image. Drag the arrow over the image box to show parts of both current and previous image.