Create Images from Text Prompts using Replicate

This n8n workflow demonstrates how to generate images from text prompts using the Replicate API, specifically leveraging the heranathalie model. It provides a basic structure to initiate an image generation request and then retrieve the results once the process is complete.

What it does

This workflow simplifies the process of interacting with Replicate for image generation:

1. Manual Trigger: The workflow starts when manually executed, allowing you to initiate image generation on demand.
2. Edit Fields (Set): This node is typically used to prepare or define the input data for the image generation, such as the text prompt. (Note: The provided JSON does not include specific field definitions for this node, so you would configure it with your desired prompt and other parameters).
3. HTTP Request (Replicate API Call): It makes an HTTP POST request to the Replicate API to create a new image generation job.
4. Wait: The workflow pauses for a specified duration to allow the image generation process on Replicate to complete.
5. If: This node checks the status of the image generation job.
   • True Branch: If the job is complete, it proceeds to fetch the results.
   • False Branch: If the job is not yet complete (or failed), it might loop back or handle the error. (Note: The provided JSON shows the "If" node but no subsequent connections, implying further steps for success/failure would be added here).
6. HTTP Request (Get Image Result): If the job is successful, this node makes another HTTP GET request to the Replicate API to retrieve the generated image(s).
7. Code: This node can be used for custom JavaScript logic, such as parsing the API response, extracting image URLs, or formatting data for subsequent actions.

Prerequisites/Requirements

• n8n Instance: A running n8n instance.
• Replicate Account: An account with Replicate and an API key.
• Replicate Model: Access to the heranathalie image generation model on Replicate.

Setup/Usage

1. Import the Workflow:
   • Copy the provided JSON code.
   • In your n8n instance, go to "Workflows".
   • Click "New" and then "Import from JSON".
   • Paste the JSON code and click "Import".
2. Configure Credentials:
   • Locate the "HTTP Request" nodes.
   • You will need to configure a credential for Replicate (e.g., an "HTTP Basic Auth" or "API Key" credential) to authenticate your requests. This typically involves setting your Replicate API key in the Authorization header.
3. Define Image Prompt:
   • In the "Edit Fields" (Set) node, configure the fields to include your desired text prompt for image generation. For example, you might add a field named prompt with a value like "A futuristic city at sunset, highly detailed, cyberpunk style".
4. Configure Replicate API Endpoints:
   • In the first "HTTP Request" node, ensure the URL is set to the Replicate API endpoint for creating predictions (e.g., https://api.replicate.com/v1/predictions).
   • The request body should include the model ID (e.g., heranathalie/latent-diffusion) and your prompt.
   • In the second "HTTP Request" node (after the "If" node), the URL should be dynamic, using the id of the prediction returned by the first API call to fetch the prediction details (e.g., https://api.replicate.com/v1/predictions/{{$json.id}}).
5. Adjust Wait Time:
   • Modify the "Wait" node's duration based on how long the heranathalie model typically takes to generate images.
6. Add Success/Failure Logic:
   • Expand the "If" node's branches to handle successful image generation (e.g., save the image URL to a database, post to Slack) and potential failures (e.g., send an error notification).
7. Execute the Workflow:
   • Click the "Execute Workflow" button on the "Manual Trigger" node to run the workflow.

This workflow provides a robust starting point for integrating Replicate's image generation capabilities into your automated processes.