VoyageAI | Agno

VoyageAI Embedder embeds documents using VoyageAI's embedding models, which are specifically designed for high-performance text embeddings.

Parameter	Type	Description	Default
`id`	`str`	The model ID to use for embeddings	`"voyage-2"`
`dimensions`	`int`	Output dimensions of the embedding	`1024`
`request_params`	`Optional[Dict[str, Any]]`	Additional parameters for embedding requests	`None`
`api_key`	`Optional[str]`	VoyageAI API key; when unset, the VoyageAI SDK reads `VOYAGE_API_KEY`	`None`
`base_url`	`str`	Declared field, currently not forwarded to either client; configure a custom endpoint with `client_params` or an injected client	`"https://api.voyageai.com/v1/embeddings"`
`max_retries`	`Optional[int]`	Maximum number of retry attempts	`None`
`timeout`	`Optional[float]`	Request timeout in seconds	`None`
`client_params`	`Optional[Dict[str, Any]]`	Additional parameters for client initialization	`None`
`voyage_client`	`Optional[Client]`	Pre-configured VoyageAI client	`None`
`async_client`	`Optional[AsyncClient]`	Pre-configured async VoyageAI client	`None`
`enable_batch`	`bool`	Embed multiple texts per API call when adding content to vector databases	`False`
`batch_size`	`int`	Number of texts per batch when `enable_batch` is `True`	`100`
`dimensions` records the expected vector width; it is not forwarded and does not resize embeddings. For models that support configurable widths, set `request_params={"output_dimension": ...}` and matching `dimensions`. A custom SDK endpoint can be supplied with `client_params={"base_url": "https://your-host/v1"}` or an injected client.