Text-to-Image
AI-generatedSummary
Generate one or more images from a text prompt using a selected Seedream model with customizable aspect ratio and optional negative prompts to exclude elements.
Inputs
- Model (required) — Choose the Seedream model for image generation, defaulting to Seedream 5.0 Lite.
- Prompt (required) — Provide the text prompt describing the desired image content.
- Aspect Ratio (required) — Select the aspect ratio of the generated image, such as Square HD or various landscape and portrait options.
- Negative Prompt — Specify elements to avoid in the generated image.
- Number of Images (required) — Set the number of images to generate, between 1 and 4.
- Seed — Set the random seed for generation; use 0 to generate a random seed each time.
- Reply URL — Optional webhook URL called upon task completion.
- Reply Ref — Custom reference passed back in the webhook callback.
- Wait for Completion (required) — Choose whether to wait synchronously for the task to complete (with polling every 3 seconds and a timeout of 5 minutes).
Output shape
a JSON array where each item corresponds to one input and contains the Kie.ai API response for the generation task, including URLs to the generated images when completed.
If 'Wait for Completion' is true, the node polls until the generation task succeeds or fails, returning the final task data including image URLs. Otherwise, it returns immediately with the task creation response, which includes a taskId that can be used with 'Query Task Status' operation to check progress later.
Examples
Example 1: Generate a single square high-definition image of a sunset with no undesirable elements.
Set 'Model' to 'Seedream 5.0 Lite', 'Prompt' to 'A beautiful sunset over the ocean', 'Aspect Ratio' to 'Square HD', 'Negative Prompt' to exclude elements, 'Number of Images' to 1, keep default seed, and 'Wait for Completion' to true.