n8n Integration

Overview

n8n is a workflow automation platform that enables you to connect apps and automate workflows. With its visual, node-based interface, you can build complex automations without extensive coding knowledge. n8n supports both cloud-hosted and self-hosted deployments, giving you flexibility in how you manage your automation infrastructure.

The Cloudinary n8n integration allows you to connect Cloudinary's capabilities with over a thousand apps, data sources, services, and n8n's built-in AI features. Using Cloudinary as the central source of truth for media files within your organization enables efficient management of large volumes of media.

In your n8n workflows, you can use instances of the Cloudinary node to:

• Upload assets from file data or a URL
• Build optimized transformation and delivery URLs for images and videos, including crop, resize, optimize, format conversion, and custom transformation strings
• Generate video players
• Generate images with AI from a text prompt or from reference images
• Manage assets by searching, retrieving, deleting, and updating tags and structured metadata
• Retrieve a cloud's tag list and structured metadata field definitions

n8n integration video tutorial

This video tutorial demonstrates how to install and use the Cloudinary node in your n8n automation:

Tutorial contents

Get started

Before you install the Cloudinary node, make sure you have a Cloudinary account. You can start by signing up for a Free plan. When your requirements grow, you can upgrade to a plan that best fits your needs.

Once you have your Cloudinary account, install the Cloudinary node from the nodes panel in n8n. Search for Cloudinary and click Install node.

Add credentials

Connect the Cloudinary node to your Cloudinary product environment by adding a new credential.

1. From the Cloudinary node panel, select one of the actions and click the Credential to connect with field.
2. Click Create new credential.
3. In the Cloudinary account dialog, enter your Cloud Name, API Key and API Secret.
   Tip

   Find your product environment's Cloud Name on the Dashboard and your API Key and Secret in the API Keys page of the Console Settings.
4. Save the settings. You can now use these credentials for other Cloudinary actions in your automations.

Supported operations

The Cloudinary node groups its operations by resource:

Generate images with AI

Use the Generate resource of the Cloudinary node to create new images with the Image Generation add-on. By default, generated images are saved to your product environment as managed assets, ready for delivery and transformation. To get a short-lived URL instead, set the storage option to Temporary.

The Generate resource includes these operations:

The Generate resource is also available when you use the Cloudinary node as a tool for an n8n AI Agent.

Select a model

Choose the model that generates your images in one of these ways:

• Default: Uses the node's default model.
• Auto: Cloudinary selects a model based on your request.
• By Family and Tier: Select a model family and a tier. The Standard tier is faster and uses fewer credits. The Premium tier gives the highest quality and uses more credits.
• By Model ID: Select a specific model. Text-to-image and image-to-image operations have separate model lists. Models with an -edit suffix are available only for generating images from reference images. To use a new model that isn't listed yet, enter its ID as an expression in the Model ID field.

For the list of available models, see Supported models.

Synchronous and asynchronous generation

By default, the node waits for the generation to complete and returns the generated images.

To generate asynchronously, turn on Options > Async. The node returns a task_id instead of the images. To get the results, either:

• Pass the task_id to a Get Generation Task operation, for example after a Wait node.
• Specify a Notification URL to receive the results as a webhook.

The node doesn't poll for task results on its own. If a task's status is failed, the Get Generation Task operation throws an error, so your workflow's Continue On Fail and error output settings apply.

Generation output

The node outputs one item for each generated image, including the public_id, asset_id, secure_url, format, width, height, model, seed, and request_id of the image. The output also includes limits, showing the credits used and remaining, and any notices from the API. To return the raw API response instead, turn off Simplify.

Credit usage varies by model and resolution. To track your usage, check the limits field in the output or the Image Generation add-on page in the Console.

If a request fails, the error message includes the API's error code, for example MG_00704, and the request_id. Requests must use image sizes between 64 and 4096 pixels and HTTPS URLs for reference images.

Example: Generate and optimize an image

You can pass the generated image directly to other Cloudinary operations in your workflow:

1. Add a Cloudinary node with the Generate Image From Text operation, and enter your prompt.
2. Add a Cloudinary node with a Transform operation, such as Image: Optimize or Image: Resize, and set the public ID to {{ $json.public_id }}. The node returns an optimized delivery URL for the generated image.
3. To add tags or metadata to the generated image, add a Cloudinary node with an asset operation and set the asset ID to {{ $json.asset_id }}.

For asynchronous generation, add a Wait node and a Get Generation Task operation between the generate and transformation steps.

Cloudinary Professional Services configuration support

The n8n integration is part of Cloudinary's Standard Professional Services integration configuration package.

If you need assistance setting up this integration and you're on an Enterprise plan with a professional services package in your contract, Cloudinary's professional services can help you set it up as part of your contracted package.

Learn more: Cloudinary Professional Services packages