> ## Documentation Index
> Fetch the complete documentation index at: https://docs.squared.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Weaviate

> Connect AI Squared to Weaviate to sync vector data to your Weaviate collections.

## Connect AI Squared to Weaviate

This guide will help you configure the Weaviate Connector in AI Squared to access and transfer data to your Weaviate collection.

### Prerequisites

Before proceeding, ensure you have your API URL and API key.

## Step-by-Step Guide to Connect to your Weaviate collection

## Step 1: Navigate to Weaviate

Start by logging into your Weaviate account.

1. Sign in to your Weaviate account at [Weaviate Cloud Console](https://console.weaviate.cloud/)

<Frame>
  <img src="https://res.cloudinary.com/dspflukeu/image/upload/Multiwoven/connectors/Weaviate/Weaviate_Login_s5pzsl.png" />
</Frame>

2. Select **Clusters** from the side bar.

<Frame>
  <img src="https://res.cloudinary.com/dspflukeu/image/upload/Multiwoven/connectors/Weaviate/Weaviate_getting_started_bcdnab.png" />
</Frame>

3. Select your cluster to see more details.

<Frame>
  <img src="https://res.cloudinary.com/dspflukeu/image/upload/Multiwoven/connectors/Weaviate/Weaviate_cluster_creation_fkogk3.png" />
</Frame>

## Step 2: Locate Weaviate Configuration Details

Once in your Weaviate cluster, you'll find the necessary configuration details:

**API URL:**

* Click on the **Details** tab of your cluster.
* Copy the **Cluster URL** (API URL).

<Frame>
  <img src="https://res.cloudinary.com/dspflukeu/image/upload/Multiwoven/connectors/Weaviate/Weaviate_cluster_details_radnhh.png" />
</Frame>

**API Key:**

* Click on the **API Keys** tab.
* If you haven't created an API key before, click on **Create** to generate a new one. Make sure to copy the API Key as it is shown only once.

<Frame>
  <img src="https://res.cloudinary.com/dspflukeu/image/upload/Multiwoven/connectors/Weaviate/Weaviate_cluster_authentication_hhnmuf.png" />
</Frame>

## Step 3: Configure Weaviate Connector in Your Application

Now that you have gathered all the necessary details, enter the following information in your application:

* **API Key:** The authentication key used to access your Weaviate cluster.
* **API URL:** The endpoint where your Weaviate cluster is hosted.

## Step 4: Test the Weaviate Connection

After configuring the connector in your application:

1. Save the configuration settings.
2. Test the connection to Weaviate from your application to ensure everything is set up correctly.

By following these steps, you've successfully set up a Weaviate destination connector in AI Squared.

### Supported sync modes

| Mode             | Supported (Yes/No/Coming soon) |
| ---------------- | ------------------------------ |
| Incremental sync | YES                            |
| Full refresh     | Coming soon                    |

## Performance features

### Batch write support

The Weaviate destination connector supports efficient batch writes with the following capabilities:

* **Batch operations**: Records are written in batches using Weaviate's batch create API for optimal performance
* **Upsert support**: When an `id` field is provided, the connector generates a deterministic UUID to enable upsert behavior, preventing duplicate records
* **Automatic type coercion**: Property values are automatically coerced to match the Weaviate collection schema (text, integer, number, boolean, and date types)
* **Vector support**: Each record can include a `vector` field containing embedding arrays for semantic search and AI-powered applications

### Schema discovery

The connector automatically discovers your Weaviate collections and their property schemas during sync configuration. Each collection stream includes:

| Field        | Description                                                     |
| ------------ | --------------------------------------------------------------- |
| `id`         | Optional string identifier for upsert behavior                  |
| `vector`     | Embedding vector array (array of floats)                        |
| `properties` | Object containing collection properties with auto-coerced types |