n8n Workflow: Text-to-Image with LinkModel GPT Image 2

Generate Images From Text Prompts in n8n Using LinkModel gpt-image-2

This n8n workflow lets you create images directly from a text prompt using LinkModel’s gpt-image-2 endpoint—then it automatically polls for completion and downloads each generated image for you to view inside n8n.

What this workflow does

Built for manual execution, the workflow runs end-to-end whenever you click Execute workflow in n8n:

• Prepare inputs: sets image-generation parameters—prompt, quality (low/medium/high), and size.
• Create an image task: sends the inputs to LinkModel’s image-generation API to start the job.
• Fail fast on request issues: stops with an error if LinkModel does not accept the task request.
• Poll for status: waits briefly and then repeatedly queries LinkModel until the job succeeds, fails/cancels, or hits a 15-minute deadline.
• Process results: extracts the output image URLs from the successful response and splits them into individual items.
• Download images: downloads each generated image from its URL and returns the files in the final workflow output (as binary files). If signed URLs expire, download promptly after completion.

Use cases

• Prototype creatives quickly: iterate on prompts and quality/size settings without leaving n8n.
• SaaS operator content testing: generate image variations for feature mockups or landing-page experiments.
• Automation engineers: validate asynchronous job handling (create → poll → download) against LinkModel’s API.

Technical details

• Integrations: LinkModel API (gpt-image-2) via HTTP requests using a Bearer LinkModel API key.
• n8n nodes used (per workflow): if, set, wait, switch, split out, sticky note.
• Configuration steps: replace Bearer YOUR_LINKMODEL_API_KEY in both LinkModel HTTP requests, then update prompt/quality/size.