tPjnPddlZddlmZddlmZmZddlZddlmZGddZ y)N)Iterator)AnyFinal)UnauthorizedErrorcleZdZddededzdefdZdeeeffdZ dded eeeefd e dzd e dzd edzd edzde dzde dzdedzde deee fe jzfdZ dded eeeefd e dzd e dzd edzd edzde dzde dzdedzdeeee ffdZy) ChatClientNbase_urlapi_keytimeoutcL|jd|_||_||_y)a Initialize the ChatClient. Args: base_url (str): The base URL of the LiteLLM proxy server (e.g., "http://localhost:8000") api_key (Optional[str]): API key for authentication. If provided, it will be sent as a Bearer token. timeout (int): Request timeout in seconds (default: 600, the OpenAI SDK default, since a completion can legitimately take minutes) /N)rstrip _base_url_api_key_timeout)selfr r r s ^/home/oem/CyberBrain26-unified/.venv/lib/python3.12/site-packages/litellm/proxy/client/chat.py__init__zChatClient.__init__ s""-  returncJddi}|jrd|j|d<|S)z Get the headers for API requests, including authorization if api_key is set. Returns: Dict[str, str]: Headers to use for API requests z Content-Typezapplication/jsonzBearer Authorization)r)rheaderss r _get_headerszChatClient._get_headerss0)*<= ==)0'@GO $rmodelmessages temperaturetop_pn max_tokenspresence_penaltyfrequency_penaltyuserreturn_requestc 6|jd} ||d} ||| d<||| d<||| d<||| d<||| d<||| d <| | | d <tjd | |j| } | r| Stj} |j | j |j }|j|jS#tjj$r*}|jjdk(r t|d}~wwxYw)a Create a chat completion. Args: model (str): The model to use for completion messages (List[Dict[str, str]]): The messages to generate a completion for temperature (Optional[float]): Sampling temperature between 0 and 2 top_p (Optional[float]): Nucleus sampling parameter between 0 and 1 n (Optional[int]): Number of completions to generate max_tokens (Optional[int]): Maximum number of tokens to generate presence_penalty (Optional[float]): Presence penalty between -2.0 and 2.0 frequency_penalty (Optional[float]): Frequency penalty between -2.0 and 2.0 user (Optional[str]): Unique identifier for the end user return_request (bool): If True, returns the prepared request object instead of executing it Returns: Union[Dict[str, Any], requests.Request]: Either the completion response from the server or a prepared request object if return_request is True Raises: UnauthorizedError: If the request fails with a 401 status code requests.exceptions.RequestException: If the request fails with any other error /chat/completions)rrNrrr r!r"r#r$POST)rjson)r )rrequestsRequestrSessionsendpreparerraise_for_statusr) exceptions HTTPErrorresponse status_coder)rrrrrr r!r"r#r$r%urldatarequestsessionr3es r completionszChatClient.completions%sAH''89168&L  ""-D   !DM =DI  !!+D   ''7D# $  ((9D$ %  DL!))&#t?P?P?RY]^ N"))+ %ll7??+>+x7!%$(JJx$8E"'K % $33%$%"",, zz%%,'**  saAE/AD%2"D%D%.E/0D  D% E/ D"D%!D""D%%E,%E''E,,E/)NiX)NNNNNNNF)NNNNNNN)__name__ __module__ __qualname__strintrdictrlistfloatboolrr+r,r:rrHrrr r s    sTz  3   d38n  %)"!%)-*.$GGtCH~&GT\ G t| G : G$JG $,G!4<GDjGG c3h(** *GZ%)"!%)-*.OOtCH~&OT\ O t| O : O$JO $,O!4<ODjO $sCx. !Orr ) r)collections.abcrtypingrrr+r1rr rRrrrUs  $)ssr