Request Example

Supported Models

Response Example: Free Text

Free Text Response

Learn how to query your collections in WetroCloud for both free text and structured responses.

Query your Resources

Api Key

Wetrocloud

Welcome 👋 to the Wetrocloud Developer Documentation. Here you'll learn how to easily integrate Retrieval Augmented Generation (RAG) into your applications using the Wetrocloud API.

Introduction

To get started with Wetrocloud’s API, follow these steps to create your API key

How to Access Your API Key

Learn how to create collections, insert resources, and query them using cURL in just a few minutes.

Quickstart

Learn how to use our API keys to develop products using OpenAI SDK

OpenAI Compatibility

WetroCloud currently supports the following models:

Learn how to create collections using the Wetrocloud API

Create a Collection

This API endpoint displays a list of all your existing collections.

List collections

Learn how to insert resources using the Wetrocloud API

Insert a Resource

Learn how to query resources using the Wetrocloud API

Structured Output

Learn how to classify and structure resources using WetroCloud.

Data categorization

Learn how to delete a collection using the Wetrocloud API when they are no longer needed.

Delete a Collection

Learn how to generate free text responses using the Wetrocloud API.

Text Generation (Without RAG)

Learn how to query images in WetroCloud to retrieve information, descriptions, or answers based on the image using the WetroCloud APIs image-to-text endpoint.

Image to text (OCR)

Learn how to extract content from websites using the WetroCloud scrape endpoint.

Data extraction

Token optimization

Learn how to easily integrate Retrieval Augmented Generation (RAG) into your applications.

Understand the structure of JSON request bodies and responses, including key details for success, tokens, and identifiers.

Request and Response Formats

Learn about the resources wetrocloud API supports

Supported Resource Types

Explore the available WetroCloud API endpoints for creating, inserting, querying, and categorizing resources, with detailed response formats.

API Endpoints

This endpoint allows you to create a new collection_id.

#### Request Body

- No request body parameters required.
    

#### Response

The response will be a JSON object with the following properties:

- `collection_id` (string): The ID of the created item.
    
- `success` (boolean): Indicates whether the item creation was successful.

Create Collection

This endpoint allows you to list all your existing collections.
<Note> This list has a pagination of up to <b>20</b> items per request. `Next` and `Previous` urls will be provided on every response </Note>
#### Request Body
- No request body parameters required.
    

#### Response

The response will be a JSON object with the following properties:

- `count` (number): Total number of collections. - `next` (string): URL for the next pagination item. - `previous` (string): URL for the previous pagination item. - `results` (list): A list of all available collections.

List all Collections


This endpoint allows you to add data to a collection using an HTTP POST request.

#### Request Body

- `collection_id` (text): The ID of the collection to which the data will be added.
    
- `resource` (text): The resource to be added to the collection.
    
- `type` (text): The type of the resource being added.
    

#### Response

Upon successful addition, the response will include:

- `success` (boolean): Indicates whether the operation was successful.
    
- `tokens` (number): The number of tokens generated.
    

Insert Resource

This endpoint allows the user to make an HTTP POST request to submit a query to the specified collection.

### Request Body

- `collection_id` (text): The ID of the collection to which the query is being submitted.
    
- `request_query` (text): The query being submitted to the collection.
    
- `json_schema` (text): The JSON schema for the query response.
    
- `json_schema_rules` (text): Rules for the JSON schema.
    

### Response

The response for this request is a JSON schema with the following structure:
- `response` (object): The JSON schema for the query response.
    
- `tokens` (number): The number of tokens.
    
- `success` (boolean): Indicates whether the request was successful.

Query Resource



This endpoint allows you to create a new category by providing the necessary details in the request body.

#### Request Body

- `resource` (text): Description of the resource.
    
- `type` (text): Type of the category.
    
- `json_schema` (text): JSON schema for the category.
    
- `json_schema_rules` (text): Rules for the JSON schema.
    

#### Response

The response is a JSON object with the following properties:

- `label`: The label of the response.
    
- `tokens`: The number of tokens.
    
- `success`: Indicates the success status of the request.
    

Categorize Resource

This endpoint is used to delete a collection.

#### Request Body

- `collection_id` (text) - The ID of the collection to be deleted.
    

#### Response

- `message` - A message indicating the result of the deletion.
    
- `success` - Indicates whether the deletion was successful.

Delete Collection



This endpoint allows you to retrieve specific data from  a web page in a structured JSON format based on your predefined schema.

#### Request Body

- `website` (text): Website url to be scraped.
    
- `json_schema` (text): JSON schema for the response.
    

#### Response

The response is a JSON object with the following properties:

- `response`: The JSON object structured as per the provided schema.
    
- `tokens`: The number of tokens.
    
- `success`: Indicates the success status of the request.
    



This endpoint allows you to analyze an image and  ask questions about its content 

#### Request Body

- `image_url` (text): Image url to be analyzed.
    
- `request_query` (text): Query been used to analyze image.
    

#### Response

The response is a JSON object with the following properties:

- `response`: The result of analysis.
    
- `tokens`: The number of tokens.
    
- `success`: Indicates the success status of the request.
    

Analyze Image



generate natural language responses using open source language models

#### Request Body

- `messages` (JSON): Array of messages.
    
- `model` (text): Identifier of language model to use (e.g., "llama-3.3-70b")
    

#### Response

The response is a JSON object with the following properties:

- `response`: Conversational response to the query.
    
- `tokens`: The number of tokens.
    
- `success`: Indicates the success status of the request.
    

Field	Description
`collection_id`	Collection Id.
`query`	This is the prompt.
`model` (optional)	Pass in the model you want.

Model Name	Identifier	Description
LLaMA 3.3 70B	`llama-3.3-70b`	A large, general-purpose language model with 70 billion parameters.
LLaMA 3.3 70B Versatile	`llama-3.3-70b-versatile`	Optimized version of LLaMA 3.3 70B for broader tasks and better flexibility.
DeepSeek R1 Distill LLaMA	`deepseek-r1-distill-llama-70b`	Distilled version of LLaMA 70B for faster performance and specialized tasks.
GPT-4o	`gpt-4o`	A variant of GPT-4 optimized for performance and versatility.
GPT-4o Mini	`gpt-4o-mini`	Lightweight version of GPT-4o for faster response with fewer resources.
Claude sonnet 3.7	`claude-3-7-sonnet-20250219`	Upgraded version of the Claude sonnet 3.5 model

Field	Description
`response`	Conversational response to the query.
`tokens`	Number of tokens used for processing.
`success`	Indicates whether the query was successful.

Get Started

​Free Text Response

​Request Example

​Supported Models

​Response Example: Free Text

Free Text Response

Request Example

Supported Models

Response Example: Free Text