Automated Image Generation with IBM Granite Model

This n8n workflow automates the process of generating images using the IBM Granite Vision 3.3 2b model via Replicate API. Triggered manually, the workflow begins when the user clicks ‘execute’ to initiate the process. It first sets the API key required for authentication, then creates a prediction request on the Replicate platform with specific parameters such as seed, top_k, and temperature to control the image generation. After submitting the request, the workflow extracts the prediction ID to monitor the prediction status. It waits for a short interval before polling the status endpoint until the prediction succeeds. Once the prediction is complete, it processes the output, retrieving the generated image URL and other metadata, making it ready for further use or display. This workflow is ideal for automating image creation tasks for content creators, marketers, or developers integrating AI-generated visuals into their projects.