### Start listing vectors with Wrangler CLI

Source: https://developers.cloudflare.com/vectorize/best-practices/list-vectors

Initiate a new list-vectors request to get the first page of vector identifiers. Specify the index name and the desired count per page. This starts a new snapshot for pagination.

```bash
wrangler vectorize list-vectors my-index --count=1000
```

--------------------------------

### Create Cloudflare Worker Project with npm

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

Use npm to create a new Cloudflare Worker project. Select 'Hello World example', 'Worker only', 'TypeScript', and 'Yes' for Git, 'No' for deployment during setup.

```bash
npm create cloudflare@latest -- embeddings-tutorial
```

--------------------------------

### Prefix searching with range query on strings

Source: https://developers.cloudflare.com/vectorize/reference/metadata-filtering

Implement prefix searching on string metadata fields using range queries. This example matches all string values starting with 'net'. Ensure the upper bound is lexicographically the next possible prefix.

```json
{ "someKey": { "$gte": "net", "$lt": "neu" } }
```

--------------------------------

### Vectorize index creation output

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

Example output from the creation command, including the required configuration binding for wrangler.toml.

```text
🚧 Creating index: 'tutorial-index'

✅ Successfully created a new Vectorize index: 'tutorial-index'

📋 To start querying from a Worker, add the following binding configuration into 'wrangler.toml':


[[vectorize]]

binding = "VECTORIZE" # available in your Worker on env.VECTORIZE

index_name = "tutorial-index"
```

--------------------------------

### Create Cloudflare Worker Project with pnpm

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

Use pnpm to create a new Cloudflare Worker project. Select 'Hello World example', 'Worker only', 'TypeScript', and 'Yes' for Git, 'No' for deployment during setup.

```bash
pnpm create cloudflare@latest embeddings-tutorial
```

--------------------------------

### Create Cloudflare Worker Project with yarn

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

Use yarn to create a new Cloudflare Worker project. Select 'Hello World example', 'Worker only', 'TypeScript', and 'Yes' for Git, 'No' for deployment during setup.

```bash
yarn create cloudflare embeddings-tutorial
```

--------------------------------

### Example Vector Object

Source: https://developers.cloudflare.com/vectorize/reference/client-api

An example structure for a vector object, including its ID, values, and optional metadata.

```javascript
let vectorExample = {
  id: "12345",
  values: [32.4, 6.55, 11.2, 10.3, 87.9],
  metadata: {
    key: "value",
    hello: "world",
    url: "r2://bucket/some/object.json",
  },
};
```

--------------------------------

### Create a Vectorize Index

Source: https://developers.cloudflare.com/vectorize/get-started/intro

Create a new Vectorize index named 'tutorial-index' with 32 dimensions and Euclidean distance metric. Ensure wrangler version 3.71.0 or later is installed.

```bash
npx wrangler vectorize create tutorial-index --dimensions=32 --metric=euclidean
```

--------------------------------

### Create a LangChain Agent with Tools

Source: https://developers.cloudflare.com/vectorize/examples/langchain

Initializes an agent with a custom weather tool and invokes it to process a user message. Requires installing langchain and @langchain/anthropic packages.

```typescript
// First install: npm install langchain @langchain/anthropic
import { z } from "zod";
import { createAgent, tool } from "langchain";

const getWeather = tool(
  ({ city }) => `It's always sunny in ${city}!`,
  {
    name: "get_weather",
    description: "Get the weather for a given city",
    schema: z.object({
      city: z.string(),
    }),
  },
);

const agent = createAgent({
  model: "anthropic:claude-sonnet-4-6",
  tools: [getWeather],
});

console.log(
  await agent.invoke({
    messages: [{ role: "user", content: "What's the weather in Tokyo?" }],
  })
);
```

--------------------------------

### Get Vectorize Index Info (npm)

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

Use this command to retrieve details about a specific Vectorize index. Requires the index name as an argument.

```bash
npx wrangler vectorize info [NAME]
```

--------------------------------

### Query Index Results

Source: https://developers.cloudflare.com/vectorize/get-started/intro

Example JSON output from a vector similarity search, showing top matches with scores and metadata.

```json
// https://vectorize-tutorial.<YOUR_SUBDOMAIN>.workers.dev/
{
  "matches": {
    "count": 3,
    "matches": [
      {
        "id": "4",
        "score": 0.46348256,
        "values": [
          0.17, 0.29, 0.42, 0.57, 0.64, 0.38, 0.51, 0.72, 0.22, 0.85, 0.39,
          0.66, 0.74, 0.32, 0.53, 0.48, 0.21, 0.69, 0.77, 0.34, 0.8, 0.55, 0.41,
          0.29, 0.7, 0.62, 0.35, 0.68, 0.53, 0.3, 0.79, 0.49
        ],
        "metadata": {
          "url": "/products/sku/418313"
        }
      },
      {
        "id": "3",
        "score": 0.52920616,
        "values": [
          0.21, 0.33, 0.55, 0.67, 0.8, 0.22, 0.47, 0.63, 0.31, 0.74, 0.35, 0.53,
          0.68, 0.45, 0.55, 0.7, 0.28, 0.64, 0.71, 0.3, 0.77, 0.6, 0.43, 0.39,
          0.85, 0.55, 0.31, 0.69, 0.52, 0.29, 0.72, 0.48
        ],
        "metadata": {
          "url": "/products/sku/97913813"
        }
      },
      {
        "id": "2",
        "score": 0.6337869,
        "values": [
          0.14, 0.23, 0.36, 0.51, 0.62, 0.47, 0.59, 0.74, 0.33, 0.89, 0.41,
          0.53, 0.68, 0.29, 0.77, 0.45, 0.24, 0.66, 0.71, 0.34, 0.86, 0.57,
          0.62, 0.48, 0.78, 0.52, 0.37, 0.61, 0.69, 0.28, 0.8, 0.53
        ],
        "metadata": {
          "url": "/products/sku/10148191"
        }
      }
    ]
  }
}
```

--------------------------------

### Successful Response for Listing Vectorize Indexes

Source: https://developers.cloudflare.com/vectorize/vectorize-api

This is an example of a successful JSON response when listing Vectorize indexes. It includes details about the index configuration, creation, and modification timestamps.

```json
{
  "errors": [
    {
      "code": 1000,
      "message": "message",
      "documentation_url": "documentation_url",
      "source": {
        "pointer": "pointer"
      }
    }
  ],
  "messages": [
    {
      "code": 1000,
      "message": "message",
      "documentation_url": "documentation_url",
      "source": {
        "pointer": "pointer"
      }
    }
  ],
  "result": [
    {
      "config": {
        "dimensions": 768,
        "metric": "cosine"
      },
      "created_on": "2022-11-15T18:25:44.442097Z",
      "description": "This is my example index.",
      "modified_on": "2022-11-15T18:25:44.442097Z",
      "name": "example-index"
    }
  ],
  "success": true
}
```

--------------------------------

### Get Vectorize Index Info (pnpm)

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

Use this command to retrieve details about a specific Vectorize index. Requires the index name as an argument.

```bash
pnpm wrangler vectorize info [NAME]
```

--------------------------------

### Get Vectorize Index Info using Wrangler CLI

Source: https://developers.cloudflare.com/vectorize/reference/client-api

Retrieve additional details about your Vectorize index. Requires Wrangler v3.71.0 or later.

```bash
wrangler vectorize info <index-name>
```

--------------------------------

### Get Vectorize Index Info (yarn)

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

Use this command to retrieve details about a specific Vectorize index. Requires the index name as an argument.

```bash
yarn vectorize info [NAME]
```

--------------------------------

### Navigate to project directory

Source: https://developers.cloudflare.com/vectorize/get-started/embeddings

Change into the newly created project directory.

```bash
cd embeddings-tutorial
```

--------------------------------

### Vectorize Index Creation Output and Configuration

Source: https://developers.cloudflare.com/vectorize/get-started/intro

This output confirms the successful creation of the 'tutorial-index' and provides the necessary binding configuration to be added to 'wrangler.toml' for Worker access.

```bash
🚧 Creating index: 'tutorial-index'
✅ Successfully created a new Vectorize index: 'tutorial-index'
📋 To start querying from a Worker, add the following binding configuration into 'wrangler.toml':

[[vectorize]]
binding = "VECTORIZE" # available in your Worker on env.VECTORIZE
index_name = "tutorial-index"
```

--------------------------------

### Navigate to Worker Project Directory

Source: https://developers.cloudflare.com/vectorize/get-started/intro

Change the current directory to the 'vectorize-tutorial' project folder.

```bash
cd vectorize-tutorial
```

--------------------------------

### GET vectorize get

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

Retrieves details for a specific Vectorize index by name.

```APIDOC
## GET vectorize get

### Description
Get a Vectorize index by name.

### Endpoint
wrangler vectorize get [NAME]

### Parameters
#### Path Parameters
- **NAME** (string) - Required - The name of the Vectorize index.

#### Query Parameters
- **--json** (boolean) - Optional - Return output as JSON
- **--deprecated-v1** (boolean) - Optional - Fetch a deprecated V1 Vectorize index.
```

--------------------------------

### Create Cloudflare Agents Starter Project

Source: https://developers.cloudflare.com/vectorize/examples/agents

Use this command to create a new Cloudflare Agents project. No API keys are required as it defaults to Workers AI.

```bash
npx create-cloudflare@latest --template cloudflare/agents-starter
```

```bash
cd agents-starter && npm install
```

```bash
npm run dev
```

--------------------------------

### Vectorize Get Vectors API

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

Get vectors from a Vectorize index by providing their identifiers.

```APIDOC
## GET /vectorize/get-vectors

### Description
Get vectors from a Vectorize index.

### Method
GET

### Endpoint
`/vectorize/get-vectors`

### Parameters
#### Path Parameters
- **[NAME]** (string) - Required - The name of the Vectorize index.

#### Query Parameters
- **--ids** (string) - Required - Vector identifiers to be fetched from the Vectorize Index. Example: `--ids a 'b' 1 '2'`

### Global Flags
- **--v** (boolean) - Alias: --version - Show version number
- **--cwd** (string) - Run as if Wrangler was started in the specified directory instead of the current working directory
- **--config** (string) - Alias: --c - Path to Wrangler configuration file
- **--env** (string) - Alias: --e - Environment to use for operations, and for selecting .env and .dev.vars files
- **--env-file** (string) - Path to an .env file to load - can be specified multiple times - values from earlier files are overridden by values in later files
- **--experimental-provision** (boolean) - Aliases: --x-provision - Default: true - Experimental: Enable automatic resource provisioning
- **--experimental-auto-create** (boolean) - Alias: --x-auto-create - Default: true - Automatically provision draft bindings with new resources
```

--------------------------------

### Create an index with Wrangler CLI

Source: https://developers.cloudflare.com/vectorize/best-practices/create-indexes

Use the wrangler CLI to initialize a new index with specified dimensions and distance metrics.

```bash
npx wrangler vectorize create your-index-name --dimensions=NUM_DIMENSIONS --metric=SELECTED_METRIC
```

```bash
npx wrangler vectorize create your-index-name --dimensions=768 --metric=cosine
```

--------------------------------

### Initialize a Cloudflare Worker project

Source: https://developers.cloudflare.com/vectorize/get-started/embeddings

Commands to create a new project using different package managers.

```bash
npm create cloudflare@latest -- embeddings-tutorial
```

```bash
yarn create cloudflare embeddings-tutorial
```

```bash
pnpm create cloudflare@latest embeddings-tutorial
```

--------------------------------

### Get a Vectorize Index

Source: https://developers.cloudflare.com/vectorize/reference/wrangler-commands

Use the `vectorize get` command to retrieve details about a specific Vectorize index by its name.

```APIDOC
## GET /vectorize/get

### Description
Retrieves details of a Vectorize index by its name.

### Method
GET

### Endpoint
`/vectorize/get [NAME]`

### Parameters
#### Path Parameters
- **[NAME]** (string) - Required - The name of the Vectorize index.

#### Query Parameters
- **--json** (boolean) - Optional - Return output as JSON (default: false).
- **--deprecated-v1** (boolean) - Optional - Fetch a deprecated V1 Vectorize index. This must be enabled if the index was created with V1 option (default: false).

### Global Flags
- **--v** (boolean) - Optional - Show version number (alias: --version).
- **--cwd** (string) - Optional - Run as if Wrangler was started in the specified directory instead of the current working directory.
- **--config** (string) - Optional - Path to Wrangler configuration file (alias: --c).
- **--env** (string) - Optional - Environment to use for operations, and for selecting .env and .dev.vars files (alias: --e).
- **--env-file** (string) - Optional - Path to an .env file to load - can be specified multiple times - values from earlier files are overridden by values in later files.
- **--experimental-provision** (boolean) - Optional - Experimental: Enable automatic resource provisioning (aliases: --x-provision, default: true).
- **--experimental-auto-create** (boolean) - Optional - Automatically provision draft bindings with new resources (alias: --x-auto-create, default: true).

### Request Example
```bash
npx wrangler vectorize get my-index
```

### Response
#### Success Response (200)
- **name** (string) - The name of the index.
- **id** (string) - The unique identifier of the index.
- **created_at** (string) - The timestamp when the index was created.
- **dimensions** (number) - The dimension size of the index.
- **metric** (string) - The distance metric used by the index.
- **description** (string) - The description of the index.

#### Response Example
```json
{
  "name": "my-index",
  "id": "a1b2c3d4-e5f6-7890-1234-567890abcdef",
  "created_at": "2023-10-27T10:00:00Z",
  "dimensions": 1536,
  "metric": "cosine",
  "description": "My embeddings index"
}
```
```

--------------------------------

### List all Vectorize indexes using npm

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

Execute this command to view a list of all Vectorize indexes associated with your account. No arguments are required.

```bash
npx wrangler vectorize list
```

--------------------------------

### Vector query response example

Source: https://developers.cloudflare.com/vectorize/best-practices/query-vectors

Example response structure for a vector query using the `cosine` distance metric. The response includes a count of matches and a list of matching vectors with their scores and IDs.

```json
{
  "count": 5,
  "matches": [
    { "score": 0.999909486, "id": "5" },
    { "score": 0.789848214, "id": "4" },
    { "score": 0.720476967, "id": "4444" },
    { "score": 0.463884663, "id": "6" },
    { "score": 0.378282232, "id": "1" }
  ]
}
```

--------------------------------

### Initialize a new Workers project

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

Create a new project directory with the necessary boilerplate for a TypeScript-based Worker.

```bash
npm create cloudflare@latest -- vectorize-tutorial
```

```bash
yarn create cloudflare vectorize-tutorial
```

```bash
pnpm create cloudflare@latest vectorize-tutorial
```

--------------------------------

### Example NDJSON Formatted File

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

This is an example of newline-delimited JSON (NDJSON) format for bulk uploading vector embeddings. Each vector must be on a new line and include a unique string `id`, `values`, and optional `metadata`.

```json
{ "id": "4444", "values": [175.1, 167.1, 129.9], "metadata": {"url": "/products/sku/918318313"}}
{ "id": "5555", "values": [158.8, 116.7, 311.4], "metadata": {"url": "/products/sku/183183183"}}
{ "id": "6666", "values": [113.2, 67.5, 11.2], "metadata": {"url": "/products/sku/717313811"}}

```

--------------------------------

### Initialize a project non-interactively

Source: https://developers.cloudflare.com/vectorize/get-started/embeddings

Use the CI environment variable to skip interactive prompts during project creation.

```bash
CI=true npm create cloudflare@latest embeddings-tutorial --type=simple --git --ts --deploy=false
```

--------------------------------

### Advanced vector query response example

Source: https://developers.cloudflare.com/vectorize/best-practices/query-vectors

Example response for an advanced vector query including vector values and metadata. This response is based on the `cosine` distance metric and includes a single match due to `topK: 1`.

```json
{
  "count": 1,
  "matches": [
    {
      "score": 0.999909486,
      "id": "5",
      "values": [58.79999923706055, 6.699999809265137, 3.4000000953674316, ...],
      "metadata": { "url": "/products/sku/55519183" }
    }
  ]
}
```

--------------------------------

### GET vectorize list-vectors

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

Lists vector identifiers within a specified Vectorize index.

```APIDOC
## GET vectorize list-vectors

### Description
List vector identifiers in a Vectorize index.

### Endpoint
wrangler vectorize list-vectors [NAME]

### Parameters
#### Path Parameters
- **NAME** (string) - Required - The name of the Vectorize index.

#### Query Parameters
- **--count** (number) - Optional - Maximum number of vectors to return (1-1000)
- **--cursor** (string) - Optional - Cursor for pagination to get the next page of results
- **--json** (boolean) - Optional - Return output as JSON
```

--------------------------------

### Get Vectorize Index Information

Source: https://developers.cloudflare.com/vectorize/reference/wrangler-commands

Retrieves detailed information about a specified Vectorize index.

```APIDOC
## GET /vectorize/info

### Description
Gets additional details about the specified Vectorize index.

### Method
`GET` (via CLI command)

### Endpoint
`wrangler vectorize info [NAME]`

### Parameters
#### Path Parameters
- **[NAME]** (string) - Required - The name of the Vectorize index.

#### Query Parameters
- **--json** (boolean) - Optional - Return output as JSON (default: false)

#### Global Flags
- **--v** (boolean) - alias: --version - Show version number
- **--cwd** (string) - Run as if Wrangler was started in the specified directory instead of the current working directory
- **--config** (string) - alias: --c - Path to Wrangler configuration file
- **--env** (string) - alias: --e - Environment to use for operations, and for selecting .env and .dev.vars files
- **--env-file** (string) - Path to an .env file to load - can be specified multiple times - values from earlier files are overridden by values in later files
- **--experimental-provision** (boolean) - aliases: --x-provision - Experimental: Enable automatic resource provisioning (default: true)
- **--experimental-auto-create** (boolean) - alias: --x-auto-create - Automatically provision draft bindings with new resources (default: true)
```

--------------------------------

### Create another String Metadata Index

Source: https://developers.cloudflare.com/vectorize/reference/metadata-filtering

Create a metadata index for a string property named 'streaming_platform' on an existing Vectorize index.

```bash
npx wrangler vectorize create-metadata-index tutorial-index --property-name=streaming_platform --type=string
```

--------------------------------

### Example Vector Embedding Array

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

A representation of a 768-dimension vector embedding for a text input.

```json
[-0.019273685291409492,-0.01913292706012726,<764 dimensions here>,0.0007094172760844231,0.043409910053014755]
```

--------------------------------

### Initialize Cloudflare Worker Non-Interactively

Source: https://developers.cloudflare.com/vectorize/get-started/intro

Create a basic 'Hello World' Cloudflare Worker project non-interactively using environment variables.

```bash
CI=true npm create cloudflare@latest vectorize-tutorial --type=simple --git --ts --deploy=false
```

--------------------------------

### Get Index Info (Wrangler CLI)

Source: https://developers.cloudflare.com/vectorize/reference/client-api

Retrieves additional details about a specified Vectorize index.

```APIDOC
## Get Index Info

### Description
Get additional details about the index.

### Method
`wrangler vectorize info`

### Endpoint
`wrangler vectorize info <index-name>`

### Parameters
#### Path Parameters
- **`<index-name>`** (string) - Required - The name of your Vectorize index

### Request Example
```bash
wrangler vectorize info my-index
```

### Response Example
(Response structure not provided in source text.)
```

--------------------------------

### Get Vectors by ID

Source: https://developers.cloudflare.com/vectorize/reference/client-api

Retrieves specified vectors from the index by their IDs, including their values and metadata.

```APIDOC
## Get Vectors by ID

### Description
Retrieves the specified vectors by their ID, including values and metadata.

### Method
GET (conceptual, as this is a Worker API call)

### Endpoint
`env.YOUR_INDEX.getByIds(ids)`

### Parameters
#### Query Parameters
- **ids** (Array<string>) - Required - An array of vector IDs to retrieve.

### Request Example
```javascript
let ids = ["11", "22", "33", "44"];
const vectors = await env.YOUR_INDEX.getByIds(ids);
```

### Response
#### Success Response (200)
- **vectors** (Array<Object>) - An array of vector objects, each containing `id`, `values`, and `metadata`.
  - **id** (string) - The ID of the vector.
  - **values** (Array<number>) - The values of the vector.
  - **metadata** (Object) - The metadata associated with the vector.

#### Response Example
```json
{
  "vectors": [
    {
      "id": "11",
      "values": [1.1, 2.2, 3.3],
      "metadata": {"source": "doc1"}
    },
    {
      "id": "22",
      "values": [4.4, 5.5, 6.6],
      "metadata": {"source": "doc2"}
    }
  ]
}
```
```

--------------------------------

### Create a String Metadata Index

Source: https://developers.cloudflare.com/vectorize/reference/metadata-filtering

Create a metadata index for a string property named 'url' on an existing Vectorize index.

```bash
npx wrangler vectorize create-metadata-index tutorial-index --property-name=url --type=string
```

--------------------------------

### List all Vectorize indexes using pnpm

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

Execute this command to view a list of all Vectorize indexes associated with your account. No arguments are required.

```bash
pnpm wrangler vectorize list
```

--------------------------------

### Get vectors from a Vectorize index

Source: https://developers.cloudflare.com/vectorize/reference/wrangler-commands

Use these commands to retrieve specific vectors by their identifiers from a Vectorize index.

```bash
npx wrangler vectorize get-vectors [NAME]
```

```bash
pnpm wrangler vectorize get-vectors [NAME]
```

```bash
yarn wrangler vectorize get-vectors [NAME]
```

--------------------------------

### Create Vectorize Metadata Index (npm)

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

Enable metadata filtering on a specified property for a Vectorize index. Requires the index name and property details.

```bash
npx wrangler vectorize create-metadata-index [NAME]
```

--------------------------------

### Insert vector with metadata

Source: https://developers.cloudflare.com/vectorize/best-practices/insert-vectors

Example of a vector object structure containing an ID, vector values, and associated metadata.

```javascript
{ id: '1', values: [32.4, 74.1, 3.2, ...], metadata: { path: 'r2://bucket-name/path/to/image.png', format: 'png', category: 'profile_image' } }
```

--------------------------------

### Workers AI Embedding Generation

Source: https://developers.cloudflare.com/vectorize/best-practices/query-vectors

Example of generating embeddings using Workers AI and preparing them for Vectorize queries.

```APIDOC
## Workers AI Integration

### Description
Demonstrates how to generate embeddings using a Workers AI text embedding model and then use these embeddings to query a Vectorize index.

### Request Example (Embedding Generation)
```javascript
interface EmbeddingResponse {
  shape: number[];
  data: number[][];
}

let userQuery = "a query from a user or service";
const queryVector: EmbeddingResponse = await env.AI.run(
  "@cf/baai/bge-base-en-v1.5",
  {
    text: [userQuery],
  },
);

// To query Vectorize, use the vector data:
let matches = await env.TEXT_EMBEDDINGS.query(queryVector.data[0], { topK: 1 });
```

### Notes
- When passing embeddings generated by Workers AI to the `query()` method, ensure you pass only the vector embedding itself from the `.data` sub-object (e.g., `queryVector.data[0]`).
- Passing the entire `queryVector` object or `queryVector.data` will result in an error.
```

--------------------------------

### Create a Vectorize index

Source: https://developers.cloudflare.com/vectorize/get-started/embeddings

Command to create a new index and the resulting configuration output.

```bash
npx wrangler vectorize create embeddings-index --dimensions=768 --metric=cosine
```

```toml
✅ Successfully created index 'embeddings-index'


[[vectorize]]
binding = "VECTORIZE" # available in your Worker on env.VECTORIZE
index_name = "embeddings-index"
```

--------------------------------

### Define Vector Object in TypeScript

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

Example structure of a vector object including ID, embedding values, and optional metadata.

```typescript
let vectorExample = {

  id: "12345",

  values: [32.4, 6.55, 11.2, 10.3, 87.9],

  metadata: {

    key: "value",

    hello: "world",

    url: "r2://bucket/some/object.json",

  },

};
```

--------------------------------

### Create Cloudflare Worker Project with npm

Source: https://developers.cloudflare.com/vectorize/get-started/intro

Use this command to create a new Cloudflare Worker project named 'vectorize-tutorial' using npm.

```bash
npm create cloudflare@latest -- vectorize-tutorial
```

--------------------------------

### Create metadata index

Source: https://developers.cloudflare.com/vectorize/get-started/intro

Enable filtering on specific metadata fields by creating a metadata index.

```bash
npx wrangler vectorize create-metadata-index tutorial-index --property-name=url --type=string
```

```text
📋 Creating metadata index...
✅ Successfully enqueued metadata index creation request. Mutation changeset identifier: xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx.
```

--------------------------------

### Create a Vectorize Index

Source: https://developers.cloudflare.com/vectorize/reference/wrangler-commands

Use the `vectorize create` command to provision a new Vectorize index. You can specify dimensions, metric, description, and other configuration options.

```APIDOC
## POST /vectorize/create

### Description
Creates a new Vectorize index.

### Method
POST

### Endpoint
`/vectorize/create [NAME]`

### Parameters
#### Path Parameters
- **[NAME]** (string) - Required - The name of the Vectorize index to create (must be unique).

#### Query Parameters
- **--dimensions** (number) - Optional - The dimension size to configure this index for, based on the output dimensions of your ML model.
- **--metric** (string) - Optional - The distance metric to use for searching within the index.
- **--preset** (string) - Optional - The name of a preset representing an embeddings model: Vectorize will configure the dimensions and distance metric for you when provided.
- **--description** (string) - Optional - An optional description for this index.
- **--json** (boolean) - Optional - Return output as JSON (default: false).
- **--deprecated-v1** (boolean) - Optional - Create a deprecated Vectorize V1 index. This is not recommended and indexes created with this option need all other Vectorize operations to have this option enabled (default: false).
- **--use-remote** (boolean) - Optional - Use a remote binding when adding the newly created resource to your config.
- **--update-config** (boolean) - Optional - Automatically update your config file with the newly added resource.
- **--binding** (string) - Optional - The binding name of this resource in your Worker.

### Global Flags
- **--v** (boolean) - Optional - Show version number (alias: --version).
- **--cwd** (string) - Optional - Run as if Wrangler was started in the specified directory instead of the current working directory.
- **--config** (string) - Optional - Path to Wrangler configuration file (alias: --c).
- **--env** (string) - Optional - Environment to use for operations, and for selecting .env and .dev.vars files (alias: --e).
- **--env-file** (string) - Optional - Path to an .env file to load - can be specified multiple times - values from earlier files are overridden by values in later files.
- **--experimental-provision** (boolean) - Optional - Experimental: Enable automatic resource provisioning (aliases: --x-provision, default: true).
- **--experimental-auto-create** (boolean) - Optional - Automatically provision draft bindings with new resources (alias: --x-auto-create, default: true).

### Request Example
```bash
npx wrangler vectorize create my-index --dimensions 1536 --metric cosine --description "My embeddings index"
```

### Response
#### Success Response (200)
- **name** (string) - The name of the created index.
- **id** (string) - The unique identifier of the index.
- **created_at** (string) - The timestamp when the index was created.
- **dimensions** (number) - The dimension size of the index.
- **metric** (string) - The distance metric used by the index.
- **description** (string) - The description of the index.

#### Response Example
```json
{
  "name": "my-index",
  "id": "a1b2c3d4-e5f6-7890-1234-567890abcdef",
  "created_at": "2023-10-27T10:00:00Z",
  "dimensions": 1536,
  "metric": "cosine",
  "description": "My embeddings index"
}
```
```

--------------------------------

### Initialize OpenAI Client and Generate Embeddings

Source: https://developers.cloudflare.com/vectorize/best-practices/query-vectors

Initialize the OpenAI client with your API key and generate a vector embedding for a given query. Ensure you have the `YOUR_OPENAPI_KEY` environment variable set.

```javascript
const openai = new OpenAI({ apiKey: env.YOUR_OPENAPI_KEY });


let userQuery = "a query from a user or service";


let embeddingResponse = await openai.embeddings.create({
  input: userQuery,
  model: "text-embedding-ada-002",
});
```

--------------------------------

### Get a Vectorize index by name using yarn

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

Use this command to retrieve details about a specific Vectorize index by its name. Ensure you have the index name available.

```bash
yarn wrangler vectorize get [NAME]
```

--------------------------------

### List Vectorize Metadata Indexes (npm)

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

List all metadata properties on which filtering is enabled for a given Vectorize index. Requires the index name.

```bash
npx wrangler vectorize list-metadata-index [NAME]
```

--------------------------------

### Get a Vectorize index by name using pnpm

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

Use this command to retrieve details about a specific Vectorize index by its name. Ensure you have the index name available.

```bash
pnpm wrangler vectorize get [NAME]
```

--------------------------------

### Create a Vectorize Index (Legacy V1)

Source: https://developers.cloudflare.com/vectorize/reference/metadata-filtering

Use this command to create a legacy Vectorize V1 index. Note that legacy index creation will be deprecated by December 2024.

```bash
npx wrangler vectorize create tutorial-index --dimensions=32 --metric=cosine
```

--------------------------------

### Create Vectorize Metadata Index (pnpm)

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

Enable metadata filtering on a specified property for a Vectorize index. Requires the index name and property details.

```bash
pnpm wrangler vectorize create-metadata-index [NAME]
```

--------------------------------

### Get a Vectorize index by name using npm

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

Use this command to retrieve details about a specific Vectorize index by its name. Ensure you have the index name available.

```bash
npx wrangler vectorize get [NAME]
```

--------------------------------

### Query a Vectorize index without metadata filtering

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

Use the `query()` method with `topK`, `returnValues`, and `returnMetadata` to get similar vectors. Ensure `env.YOUR_INDEX` is configured.

```typescript
let queryVector: Array<number> = [54.8, 5.5, 3.1, ...];

let originalMatches = await env.YOUR_INDEX.query(queryVector, {

  topK: 3,

  returnValues: true,

  returnMetadata: 'all',

});
```

```json
{
  "count": 3,
  "matches": [
    {
      "id": "5",
      "score": 0.999909486,
      "values": [58.79999923706055, 6.699999809265137, 3.4000000953674316],
      "metadata": {
        "url": "/products/sku/55519183",
        "streaming_platform": "hbo"
      }
    },
    {
      "id": "4",
      "score": 0.789848214,
      "values": [75.0999984741211, 67.0999984741211, 29.899999618530273],
      "metadata": {
        "url": "/products/sku/418313",
        "streaming_platform": "netflix"
      }
    },
    {
      "id": "2",
      "score": 0.611976262,
      "values": [15.100000381469727, 19.200000762939453, 15.800000190734863],
      "metadata": {
        "url": "/products/sku/10148191",
        "streaming_platform": "hbo"
      }
    }
  ]
}
```

--------------------------------

### List all Vectorize indexes using yarn

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

Execute this command to view a list of all Vectorize indexes associated with your account. No arguments are required.

```bash
yarn wrangler vectorize list
```

--------------------------------

### Create a Vectorize Index using Wrangler

Source: https://developers.cloudflare.com/vectorize/reference/wrangler-commands

Use this command to create a new Vectorize index. Specify the index name and optionally configure dimensions, metric, preset, description, and binding name. Use the `--json` flag for JSON output and `--deprecated-v1` for V1 indexes.

```bash
npx wrangler vectorize create [NAME]
```

```bash
pnpm wrangler vectorize create [NAME]
```

```bash
yarn wrangler vectorize create [NAME]
```

--------------------------------

### Query Vectorize Index with OpenAI Embedding

Source: https://developers.cloudflare.com/vectorize/best-practices/query-vectors

Query a Cloudflare Vectorize index using the generated vector embedding. This example assumes you have a Vectorize index bound to `env.TEXT_EMBEDDINGS`.

```javascript
let matches = await env.TEXT_EMBEDDINGS.query(embeddingResponse.embedding[0], {
  topK: 1,
});
```

--------------------------------

### $in operator for metadata filtering

Source: https://developers.cloudflare.com/vectorize/reference/metadata-filtering

Filter vectors where a metadata field's value is present in a specified array. This example includes vectors with metadata values 'hbo' or 'netflix'.

```json
{ "someKey": { "$in": ["hbo", "netflix"] } }
```

--------------------------------

### Create Cloudflare Worker Project with pnpm

Source: https://developers.cloudflare.com/vectorize/get-started/intro

Use this command to create a new Cloudflare Worker project named 'vectorize-tutorial' using pnpm.

```bash
pnpm create cloudflare@latest vectorize-tutorial
```

--------------------------------

### Define a Counter Agent

Source: https://developers.cloudflare.com/vectorize/examples/agents

Extend the `Agent` class and use the `@callable()` decorator to expose methods that can be called by clients over WebSocket. This example shows a simple counter agent.

```typescript
import { Agent, callable } from "agents";


export class CounterAgent extends Agent {
  initialState = { count: 0 };


  @callable()
  increment() {
    this.setState({ count: this.state.count + 1 });
    return this.state.count;
  }
}
```

```typescript
import { Agent, callable } from "agents";


export class CounterAgent extends Agent<Env, { count: number }> {
  initialState = { count: 0 };


  @callable()
  increment() {
    this.setState({ count: this.state.count + 1 });
    return this.state.count;
  }
}
```

--------------------------------

### Create a Vectorize index with Python

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

Uses the requests library to send a POST request to the Cloudflare Vectorize API. Requires a valid account ID and API token.

```python
import requests


url = "https://api.cloudflare.com/client/v4/accounts/{}/vectorize/v2/indexes".format("your-account-id")


headers = {

    "Authorization": "Bearer <your-api-token>"

}


body = {

  "name": "demo-index",

  "description": "some index description",

  "config": {

    "dimensions": 1024,

    "metric": "euclidean"

  },

}


resp = requests.post(url, headers=headers, json=body)


print('Status Code:', resp.status_code)

print('Response JSON:', resp.json())
```

--------------------------------

### Authenticate with Wrangler

Source: https://developers.cloudflare.com/vectorize/get-started/intro

Log in to your Cloudflare account via the CLI to enable deployment.

```bash
npx wrangler login
```

--------------------------------

### Get vectors from a Vectorize index using yarn

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

Retrieve specific vectors from your Vectorize index by providing their IDs. This command requires the index name and a list of vector identifiers.

```bash
yarn wrangler vectorize get-vectors [NAME] --ids <ids>
```

--------------------------------

### Get vectors from a Vectorize index using pnpm

Source: https://developers.cloudflare.com/vectorize/llms-full.txt

Retrieve specific vectors from your Vectorize index by providing their IDs. This command requires the index name and a list of vector identifiers.

```bash
pnpm wrangler vectorize get-vectors [NAME] --ids <ids>
```