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

# add_texts

Adds text documents to the vector store with optional metadata.

<Tabs>
  <Tab title="Embedded">
    ```python theme={null}
    add_texts(
        texts: Iterable[str],
        metadatas: Optional[List[dict]] = None,
        *,
        ids: Optional[List[str]] = None,
        **kwargs: Any
    ) -> List[str]
    ```

    ### Parameters

    | Parameter   | Type                   | Description                                                             |
    | ----------- | ---------------------- | ----------------------------------------------------------------------- |
    | `texts`     | `Iterable[str]`        | Iterable of text strings to add                                         |
    | `metadatas` | `Optional[List[dict]]` | *(Optional)* List of metadata dictionaries for each text                |
    | `ids`       | `Optional[List[str]]`  | *(Optional)* List of IDs for the texts (auto-generated if not provided) |
    | `**kwargs`  | `Any`                  | Additional keyword arguments (currently unused)                         |

    ### Returns

    `List[str]`: List of IDs for the added texts

    ### Example Usage

    ```python theme={null}
    # Add texts with auto-generated IDs
    texts = ["Document about AI", "Document about ML", "Document about NLP"]
    ids = store.add_texts(texts)
    print(f"Added {len(ids)} documents")

    # Add texts with metadata and custom IDs
    texts = ["Python tutorial", "JavaScript guide"]
    metadatas = [
        {"language": "python", "level": "beginner"},
        {"language": "javascript", "level": "intermediate"}
    ]
    custom_ids = ["doc1", "doc2"]
    ids = store.add_texts(texts, metadatas=metadatas, ids=custom_ids)
    ```
  </Tab>

  <Tab title="Python SDK">
    ```python theme={null}
    add_texts(
        texts: Iterable[str],
        metadatas: Optional[List[dict]] = None,
        *,
        ids: Optional[List[str]] = None,
        **kwargs: Any
    ) -> List[str]
    ```

    ### Parameters

    | Parameter   | Type                   | Description                                                             |
    | ----------- | ---------------------- | ----------------------------------------------------------------------- |
    | `texts`     | `Iterable[str]`        | Iterable of text strings to add                                         |
    | `metadatas` | `Optional[List[dict]]` | *(Optional)* List of metadata dictionaries for each text                |
    | `ids`       | `Optional[List[str]]`  | *(Optional)* List of IDs for the texts (auto-generated if not provided) |
    | `**kwargs`  | `Any`                  | Additional keyword arguments                                            |

    ### Returns

    `List[str]`: List of IDs for the added texts

    <Note>The Python SDK supports passing pre-computed embeddings via `kwargs` for cases where you want to skip the embedding step.</Note>

    ### Example Usage

    ```python theme={null}
    texts = ["Document about AI", "Document about ML"]
    metadatas = [{"topic": "ai"}, {"topic": "ml"}]
    ids = store.add_texts(texts, metadatas=metadatas)
    ```
  </Tab>

  <Tab title="JS/TS">
    ```typescript theme={null}
    addTexts(
        texts: string[],
        metadatas?: Record<string, any>[],
        options?: { ids?: string[] }
    ): Promise<string[]>
    ```

    ### Parameters

    | Parameter   | Type                    | Description                                           |
    | ----------- | ----------------------- | ----------------------------------------------------- |
    | `texts`     | `string[]`              | Array of text strings to add                          |
    | `metadatas` | `Record<string, any>[]` | *(Optional)* Array of metadata objects for each text  |
    | `options`   | `object`                | *(Optional)* Options object with optional `ids` array |

    ### Returns

    `Promise<string[]>`: Array of IDs for the added texts

    <Note>The JS/TS SDK also provides `addVectors()` for adding pre-computed embedding vectors directly, bypassing the embedding step.</Note>

    ### Example Usage

    ```typescript theme={null}
    const texts = ["Document about AI", "Document about ML"];
    const metadatas = [{ topic: "ai" }, { topic: "ml" }];
    const ids = await store.addTexts(texts, metadatas);
    ```
  </Tab>
</Tabs>

### Exceptions

<AccordionGroup>
  <Accordion title="ValueError">
    * Throws if length of ids doesn't match length of texts
  </Accordion>
</AccordionGroup>

### Async

<Accordion title="Python async variants">
  The Embedded and Python SDK provide async versions of this method prefixed with `a`:

  ```python theme={null}
  # aadd_texts — async variant
  ids = await store.aadd_texts(texts, metadatas=metadatas)
  ```
</Accordion>

<Note>
  JS/TS methods are natively async — all signatures above already return `Promise<...>`. No separate async variant is needed.
</Note>