Library

Bases: BaseLibrary

Asynchronous library client for interacting with Airia Library endpoints.

Source code in airia/client/library/async_library.py

def __init__(self, request_handler: AsyncRequestHandler):
    super().__init__(request_handler)

get_models(page_number=None, page_size=None, sort_by=None, sort_direction=None, search=None, providers=None, categories=None, licenses=None, industries=None, authors=None, is_open_source=None, chat_specialized=None, industry=None, commercial_use=None, certifications=None, has_tool_support=None, has_stream_support=None, correlation_id=None) async

Asynchronously retrieve models from the Airia Library with optional filtering and pagination.

Parameters:

Name	Type	Description	Default
page_number	Optional[int]	The page number to be fetched	None
page_size	Optional[int]	The number of items per page	None
sort_by	Optional[str]	Property to sort by	None
sort_direction	Optional[str]	Direction of the sort, either "ASC" for ascending or "DESC" for descending	None
search	Optional[str]	An optional search string	None
providers	Optional[str]	Library service provider type filter	None
categories	Optional[str]	Library service model category filter	None
licenses	Optional[str]	Library service model license filter	None
industries	Optional[str]	Optional list of industries to filter by	None
authors	Optional[str]	Optional list of authors to filter by	None
is_open_source	Optional[bool]	Optional flag to filter by open source status	None
chat_specialized	Optional[bool]	Optional flag to filter by chat specialized status	None
industry	Optional[str]	Optional flag to filter by Industry	None
commercial_use	Optional[bool]	Optional flag to filter by Commercial Use	None
certifications	Optional[str]	Optional list of certifications to filter by	None
has_tool_support	Optional[bool]	Optional flag to filter by the models support for tools	None
has_stream_support	Optional[bool]	Optional flag to filter by the models support for response streaming	None
correlation_id	Optional[str]	Optional correlation ID for request tracing. If not provided, one will be generated automatically.	None

Returns:

Name	Type	Description
GetLibraryModelsResponse	GetLibraryModelsResponse	Response containing the list of models and total count

Raises:

Type	Description
AiriaAPIError	If the API request fails with details about the error.
ClientError	For other request-related errors.

Example

client = AiriaAsyncClient(api_key="your_api_key")
response = await client.library.get_models(
    search="gpt",
    providers="OpenAI",
    page_size=10
)
for model in response.models:
    print(f"Model: {model.name} (ID: {model.id})")

Source code in airia/client/library/async_library.py

async def get_models(
    self,
    page_number: Optional[int] = None,
    page_size: Optional[int] = None,
    sort_by: Optional[str] = None,
    sort_direction: Optional[str] = None,
    search: Optional[str] = None,
    providers: Optional[str] = None,
    categories: Optional[str] = None,
    licenses: Optional[str] = None,
    industries: Optional[str] = None,
    authors: Optional[str] = None,
    is_open_source: Optional[bool] = None,
    chat_specialized: Optional[bool] = None,
    industry: Optional[str] = None,
    commercial_use: Optional[bool] = None,
    certifications: Optional[str] = None,
    has_tool_support: Optional[bool] = None,
    has_stream_support: Optional[bool] = None,
    correlation_id: Optional[str] = None,
) -> GetLibraryModelsResponse:
    """
    Asynchronously retrieve models from the Airia Library with optional filtering and pagination.

    Args:
        page_number: The page number to be fetched
        page_size: The number of items per page
        sort_by: Property to sort by
        sort_direction: Direction of the sort, either "ASC" for ascending or "DESC" for descending
        search: An optional search string
        providers: Library service provider type filter
        categories: Library service model category filter
        licenses: Library service model license filter
        industries: Optional list of industries to filter by
        authors: Optional list of authors to filter by
        is_open_source: Optional flag to filter by open source status
        chat_specialized: Optional flag to filter by chat specialized status
        industry: Optional flag to filter by Industry
        commercial_use: Optional flag to filter by Commercial Use
        certifications: Optional list of certifications to filter by
        has_tool_support: Optional flag to filter by the models support for tools
        has_stream_support: Optional flag to filter by the models support for response streaming
        correlation_id: Optional correlation ID for request tracing. If not provided,
                      one will be generated automatically.

    Returns:
        GetLibraryModelsResponse: Response containing the list of models and total count

    Raises:
        AiriaAPIError: If the API request fails with details about the error.
        aiohttp.ClientError: For other request-related errors.

    Example:
        ```python
        client = AiriaAsyncClient(api_key="your_api_key")
        response = await client.library.get_models(
            search="gpt",
            providers="OpenAI",
            page_size=10
        )
        for model in response.models:
            print(f"Model: {model.name} (ID: {model.id})")
        ```
    """
    request_data = self._prepare_get_models_request(
        page_number=page_number,
        page_size=page_size,
        sort_by=sort_by,
        sort_direction=sort_direction,
        search=search,
        providers=providers,
        categories=categories,
        licenses=licenses,
        industries=industries,
        authors=authors,
        is_open_source=is_open_source,
        chat_specialized=chat_specialized,
        industry=industry,
        commercial_use=commercial_use,
        certifications=certifications,
        has_tool_support=has_tool_support,
        has_stream_support=has_stream_support,
        correlation_id=correlation_id,
    )

    resp = await self._request_handler.make_request("GET", request_data)
    return GetLibraryModelsResponse(**resp)