Save yourself the work of writing custom integrations for Cloudinary and Google Vertex AI and use n8n instead. Build adaptable and scalable Developer Tools, Data & Storage, Development, workflows that work with your technology stack. All within a building experience you will love.

In n8n, click the "Add workflow" button in the Workflows tab to create a new workflow. Add the starting point – a trigger on when your workflow should run: an app event, a schedule, a webhook call, another workflow, an AI chat, or a manual trigger. Sometimes, the HTTP Request node might already serve as your starting point.
Upload File
Upload an asset from file data
Upload From URL
Upload an asset from URL
Compose: Combine Transformations
Build a delivery URL that chains several transformation steps in order. Outputs secure_url and a reusable transformation string.
Compose: Custom Transformation String
Build a delivery URL from a raw Cloudinary transformation string. Outputs secure_url and a reusable transformation string.
Image: Convert Format
Build a delivery URL that converts an image to another format. Outputs secure_url and a reusable transformation string.
Image: Crop
Build a delivery URL that crops an image to fixed dimensions or an aspect ratio. Outputs secure_url and a reusable transformation string.
Image: Optimize
Build a delivery URL that auto-optimizes an image (format + quality). Outputs secure_url and a reusable transformation string.
Image: Resize
Build a delivery URL that resizes an image to a width and/or height. Outputs secure_url and a reusable transformation string.
Video: Crop
Build a delivery URL that crops a video to fixed dimensions or an aspect ratio. Outputs secure_url and a reusable transformation string.
Video: Optimize
Build a delivery URL that auto-optimizes a video (format/codec + quality). Outputs secure_url and a reusable transformation string.
Video: Resize
Build a delivery URL that resizes a video to a width and/or height. Outputs secure_url and a reusable transformation string.
Video: Thumbnail
Build a delivery URL for a still image frame from a video. Outputs secure_url and a reusable transformation string.
Video: Trim
Build a delivery URL that trims a video to a start, end, and/or duration. Outputs secure_url and a reusable transformation string.
Delete Assets
Delete one or more assets by public ID
Get Asset
Get details for a single asset by asset ID
Search Assets
Search for assets using a Cloudinary search expression
Update Asset Structured Metadata
Update structured metadata for an asset by asset ID
Update Asset Tags
Update tags for an asset by asset ID
Video Player
Generate embed code and config for the Cloudinary Video Player. Outputs embed_url and a player_config JSON string.
Get Metadata Fields
Get all metadata fields definitions
Get Tags
Get all tags for a specific resource type
Update Asset Structured Metadata
Deprecated — use the Asset resource instead. Update structured metadata for an existing asset by public ID.
Update Asset Tags
Deprecated — use the Asset resource instead. Update tags for an existing asset by public ID.
Generate content
This endpoint generates content based on the input provided.
Call function
This endpoint allows for calling functions as part of the generative tasks.
Ground content
This endpoint grounds the content to ensure relevance and context.
List API errors
This endpoint retrieves potential API errors that can occur during requests.
Generate text embeddings
This endpoint generates embeddings for given text inputs.
Generate multimodal embeddings
This endpoint generates embeddings that leverage multiple modalities.
Generate and edit images
This endpoint is used for generating and editing images based on input specifications.
Use code completions
API for generating code completion suggestions.
Perform batch predictions
API for executing batch predictions on data.
Batch prediction
API for performing batch prediction.
Tune models
API for tuning machine learning models.
Tuning model parameters
API for tuning model parameters.
Rapid evaluation
API for quickly evaluating model performance.
Evaluate model performance
API for rapid evaluation of models.
Use LlamaIndex
API for accessing the LlamaIndex for retrieval-augmented generations.
Manage extensions
API for managing custom extensions.
Manage extensions
API for handling extensions.
Count tokens
API for counting tokens in text inputs.
Use reasoning engine
API for performing reasoning tasks.
Utilize reasoning engine
API for using the reasoning engine capabilities.
Use MedLM API
API for accessing medical language models.
Access MedLM
API for accessing MedLM functionality.
Generate and edit images
API for generating and editing images.
Access LlamaIndex
API for accessing LlamaIndex functionality.
Count tokens
API for counting tokens in text.
To set up Google Vertex AI integration, add the HTTP Request node to your workflow canvas and authenticate it using a generic authentication method. The HTTP Request node makes custom API calls to Google Vertex AI to query the data you need using the API endpoint URLs you provide.
See the example hereThese API endpoints were generated using n8n
n8n AI workflow transforms web scraping into an intelligent, AI-powered knowledge extraction system that uses vector embeddings to semantically analyze, chunk, store, and retrieve the most relevant API documentation from web pages. Remember to check the Google Vertex AI official documentation to get a full list of all API endpoints and verify the scraped ones!

Google Vertex AI is a unified machine learning platform that enables developers to build, deploy, and manage models efficiently. It provides a wide range of tools and services, such as AutoML and datasets, to accelerate the deployment of AI solutions.
The world's most popular workflow automation platform for technical teams including
Build complex workflows, really fast