semantic_router.encoders.cohere

docs2cohere_embed_input

def docs2cohere_embed_input(docs: list[str]) -> list[dict[str, Any]]

Convert a list of texts into Cohere's inputs format.

The Cohere embed endpoint expects one EmbedInput per embedding, each holding its own content array, so a list of N texts becomes N inputs.

Arguments:

• docs (list[str]): The texts to embed.

Returns:

``0: One embed input per text.

CohereEncoder Objects

class CohereEncoder(DenseEncoder, AsymmetricDenseMixin)

Dense encoder that uses Cohere API and SDK to embed documents. Supports text only. Requires a Cohere API key from https://dashboard.cohere.com/api-keys.

__init__

def __init__(name: str | None = None,
             cohere_api_key: str | None = None,
             score_threshold: float = 0.3)

Initialize the Cohere encoder.

Arguments:

• name (str): The name of the embedding model to use such as "embed-english-v3.0" or "embed-multilingual-v3.0".
• cohere_api_key (str): The API key for the Cohere client, can also be set via the COHERE_API_KEY environment variable.
• score_threshold (float): The threshold for the score of the embedding.

Raises:

• ValueError: If no API key is provided or found in the environment.

__call__

def __call__(docs: list[Any], **kwargs) -> list[list[float]]

Encode a list of text documents into embeddings using Cohere.

Arguments:

• docs (list[Any]): List of text documents to encode.

Returns:

list[list[float]]: List of embeddings for each document.

acall

async def acall(docs: list[Any], **kwargs) -> list[list[float]]

Encode a list of text documents into embeddings using Cohere asynchronously.

Arguments:

• docs (list[Any]): List of text documents to encode.

Returns:

list[list[float]]: List of embeddings for each document.