> ## Documentation Index
> Fetch the complete documentation index at: https://mintlify.com/run-llama/LlamaIndexTS/llms.txt
> Use this file to discover all available pages before exploring further.

# Embeddings

> Core embedding interfaces for generating vector representations of text

## Overview

The Embeddings module provides base classes and interfaces for generating vector embeddings from text. Embeddings are used to convert text into numerical vectors for semantic search and retrieval.

## BaseEmbedding

Abstract base class that all embedding implementations extend.

```typescript theme={null}
import { BaseEmbedding } from "@llamaindex/core/embeddings";
```

### Properties

<ParamField path="embedBatchSize" type="number" default={10}>
  Number of texts to embed in a single batch
</ParamField>

<ParamField path="embedInfo" type="EmbeddingInfo">
  Metadata about the embedding model

  <Expandable title="EmbeddingInfo Fields">
    <ParamField path="dimensions" type="number">
      Dimensionality of the embedding vectors
    </ParamField>

    <ParamField path="maxTokens" type="number">
      Maximum tokens the model can process
    </ParamField>

    <ParamField path="tokenizer" type="Tokenizers">
      Tokenizer used by the model
    </ParamField>
  </Expandable>
</ParamField>

### Methods

<ParamField path="getTextEmbedding" type="method">
  Get embedding for a single text string

  ```typescript theme={null}
  abstract getTextEmbedding(text: string): Promise<number[]>
  ```

  <Expandable title="Parameters">
    <ParamField path="text" type="string" required>
      The text to embed
    </ParamField>
  </Expandable>

  <Expandable title="Returns">
    <ResponseField name="embedding" type="number[]">
      The embedding vector
    </ResponseField>
  </Expandable>
</ParamField>

<ParamField path="getTextEmbeddings" type="method">
  Get embeddings for multiple texts (batch processing)

  ```typescript theme={null}
  getTextEmbeddings(texts: string[]): Promise<Array<number[]>>
  ```

  <Expandable title="Parameters">
    <ParamField path="texts" type="string[]" required>
      Array of texts to embed
    </ParamField>
  </Expandable>

  <Expandable title="Returns">
    <ResponseField name="embeddings" type="number[][]">
      Array of embedding vectors, one per input text
    </ResponseField>
  </Expandable>
</ParamField>

<ParamField path="getTextEmbeddingsBatch" type="method">
  Get embeddings for texts with batching and progress tracking

  ```typescript theme={null}
  getTextEmbeddingsBatch(
    texts: string[],
    options?: BaseEmbeddingOptions
  ): Promise<Array<number[]>>
  ```

  <Expandable title="Parameters">
    <ParamField path="texts" type="string[]" required>
      Array of texts to embed
    </ParamField>

    <ParamField path="options" type="BaseEmbeddingOptions">
      Embedding options

      <Expandable title="BaseEmbeddingOptions">
        <ParamField path="logProgress" type="boolean">
          Whether to log progress to console
        </ParamField>

        <ParamField path="progressCallback" type="(current: number, total: number) => void">
          Callback function for progress updates
        </ParamField>

        <ParamField path="logger" type="Logger">
          Custom logger instance
        </ParamField>
      </Expandable>
    </ParamField>
  </Expandable>
</ParamField>

<ParamField path="getQueryEmbedding" type="method">
  Get embedding for a query (supports multi-modal content)

  ```typescript theme={null}
  getQueryEmbedding(query: MessageContentDetail): Promise<number[] | null>
  ```

  <Expandable title="Parameters">
    <ParamField path="query" type="MessageContentDetail" required>
      Query content (text or multi-modal)
    </ParamField>
  </Expandable>

  <Expandable title="Returns">
    <ResponseField name="embedding" type="number[] | null">
      The query embedding vector, or null if text extraction fails
    </ResponseField>
  </Expandable>
</ParamField>

<ParamField path="similarity" type="method">
  Calculate similarity between two embedding vectors

  ```typescript theme={null}
  similarity(
    embedding1: number[],
    embedding2: number[],
    mode?: SimilarityType
  ): number
  ```

  <Expandable title="Parameters">
    <ParamField path="embedding1" type="number[]" required>
      First embedding vector
    </ParamField>

    <ParamField path="embedding2" type="number[]" required>
      Second embedding vector
    </ParamField>

    <ParamField path="mode" type="SimilarityType" default="DEFAULT">
      Similarity metric: `DEFAULT`, `DOT_PRODUCT`, `EUCLIDEAN`, or `COSINE`
    </ParamField>
  </Expandable>

  <Expandable title="Returns">
    <ResponseField name="similarity" type="number">
      Similarity score between the two vectors
    </ResponseField>
  </Expandable>
</ParamField>

## Usage Examples

### Basic Embedding

```typescript theme={null}
import { OpenAIEmbedding } from "@llamaindex/openai";

const embedModel = new OpenAIEmbedding({
  model: "text-embedding-3-small"
});

const embedding = await embedModel.getTextEmbedding(
  "LlamaIndex is a data framework for LLM applications"
);

console.log(embedding.length); // 1536
```

### Batch Embedding with Progress

```typescript theme={null}
const texts = [
  "First document",
  "Second document",
  "Third document"
];

const embeddings = await embedModel.getTextEmbeddingsBatch(texts, {
  logProgress: true,
  progressCallback: (current, total) => {
    console.log(`Progress: ${current}/${total}`);
  }
});

console.log(embeddings.length); // 3
```

### Embedding Nodes

The `BaseEmbedding` class extends `TransformComponent` and can embed nodes directly:

```typescript theme={null}
import { Document } from "@llamaindex/core/schema";

const documents = [
  new Document({ text: "Document 1" }),
  new Document({ text: "Document 2" })
];

// Transform adds embeddings to nodes
const embeddedNodes = await embedModel.transform(documents);

console.log(embeddedNodes[0].embedding); // number[]
```

### Similarity Calculation

```typescript theme={null}
const embedding1 = await embedModel.getTextEmbedding("cat");
const embedding2 = await embedModel.getTextEmbedding("dog");
const embedding3 = await embedModel.getTextEmbedding("car");

const similarity1 = embedModel.similarity(embedding1, embedding2);
const similarity2 = embedModel.similarity(embedding1, embedding3);

console.log(similarity1 > similarity2); // true (cat and dog are more similar)
```

## Similarity Types

```typescript theme={null}
enum SimilarityType {
  DEFAULT = "cosine",
  DOT_PRODUCT = "dot_product",
  EUCLIDEAN = "euclidean",
  COSINE = "cosine"
}
```

## Token Truncation

Embeddings automatically truncate text that exceeds the model's token limit:

```typescript theme={null}
const embedModel = new OpenAIEmbedding({
  model: "text-embedding-3-small"
});

// Set embedding info with max tokens
embedModel.embedInfo = {
  dimensions: 1536,
  maxTokens: 8191,
  tokenizer: Tokenizers.CL100K_BASE
};

// Long text will be automatically truncated
const longText = "...".repeat(10000);
const embedding = await embedModel.getTextEmbedding(longText);
```