ModelScopeInstall

Summary

Generates an image from a text prompt using a selected ModelScope image model. It submits the generation request and polls the task until it succeeds, fails, or reaches the configured timeout.

Inputs

• Model (required) — Select one of the available image models. Defaults to Qwen/Qwen-Image.
• Prompt (required) — Text describing the image to generate; it cannot be blank.
• Negative Prompt — Optional description of content to avoid; defaults to an empty string.
• Size — Image dimensions, selected from the listed sizes; defaults to 1024x1024.
• Steps — Inference steps, from 10 to 100; defaults to 30.
• Timeout (Minutes) — Maximum time to poll for task completion, from 1 to 10 minutes; defaults to 5.

Output shape

A single JSON object per input item, wrapped as an n8n output item.

On success, includes task_id, status, progress, model, prompt, negative_prompt, size, steps, images, created_at, processing_time, and attempts_used. The images field contains the service's output image URLs, or an empty array if none are returned. Missing credentials, API errors, task failure, or timeout raise an error; with Continue On Fail enabled, the item instead contains an error field.

Examples

Example 1: Generate a square image and wait up to five minutes

Select a model, enter a non-empty prompt, and use the defaults or choose another listed size, step count, and timeout.

Links

• n8n community nodes installation guide