# This file was auto-generated by Fern from our API Definition. import typing from json.decoder import JSONDecodeError from ..core.api_error import ApiError from ..core.client_wrapper import AsyncClientWrapper, SyncClientWrapper from ..core.http_response import AsyncHttpResponse, HttpResponse from ..core.pydantic_utilities import parse_obj_as from ..core.request_options import RequestOptions from ..core.serialization import convert_and_respect_annotation_metadata from ..errors.bad_request_error import BadRequestError from ..errors.forbidden_error import ForbiddenError from ..errors.internal_server_error import InternalServerError from ..errors.too_many_requests_error import TooManyRequestsError from ..errors.unprocessable_entity_error import UnprocessableEntityError from ..requests.chat_completion_request_message import ChatCompletionRequestMessageParams from ..requests.stop_configuration import StopConfigurationParams from ..types.create_chat_completion_response import CreateChatCompletionResponse from ..types.reasoning_effort import ReasoningEffort # this is used as the default value for optional parameters OMIT = typing.cast(typing.Any, ...) class RawChatClient: def __init__(self, *, client_wrapper: SyncClientWrapper): self._client_wrapper = client_wrapper def completions( self, *, messages: typing.Sequence[ChatCompletionRequestMessageParams], temperature: typing.Optional[float] = OMIT, top_p: typing.Optional[float] = OMIT, reasoning_effort: typing.Optional[ReasoningEffort] = OMIT, max_tokens: typing.Optional[int] = OMIT, stream: typing.Optional[bool] = OMIT, stop: typing.Optional[StopConfigurationParams] = OMIT, n: typing.Optional[int] = OMIT, seed: typing.Optional[int] = OMIT, frequency_penalty: typing.Optional[float] = OMIT, presence_penalty: typing.Optional[float] = OMIT, wiki_grounding: typing.Optional[bool] = OMIT, request_options: typing.Optional[RequestOptions] = None, ) -> HttpResponse[CreateChatCompletionResponse]: """ Calls Sarvam LLM API to get the chat completion. Supported model(s): `sarvam-m`. Parameters ---------- messages : typing.Sequence[ChatCompletionRequestMessageParams] A list of messages comprising the conversation so far. temperature : typing.Optional[float] What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or `top_p` but not both. top_p : typing.Optional[float] An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered. We generally recommend altering this or `temperature` but not both. reasoning_effort : typing.Optional[ReasoningEffort] max_tokens : typing.Optional[int] The maximum number of tokens that can be generated in the chat completion. stream : typing.Optional[bool] If set to true, the model response data will be streamed to the client as it is generated using [server-sent events](https://developer.mozilla.org/en-US/docs/Web/API/Server-sent_events/Using_server-sent_events#Event_stream_format). stop : typing.Optional[StopConfigurationParams] n : typing.Optional[int] How many chat completion choices to generate for each input message. Note that you will be charged based on the number of generated tokens across all of the choices. Keep `n` as `1` to minimize costs. seed : typing.Optional[int] This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same `seed` and parameters should return the same result. frequency_penalty : typing.Optional[float] Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim. presence_penalty : typing.Optional[float] Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics. wiki_grounding : typing.Optional[bool] If this parameter is enabled, then the model uses a RAG based approach to retrieve relevant chunks from Wikipedia and uses them to answer the question. This is particularly useful for queries seeking factual information. request_options : typing.Optional[RequestOptions] Request-specific configuration. Returns ------- HttpResponse[CreateChatCompletionResponse] Successful Response """ _response = self._client_wrapper.httpx_client.request( "v1/chat/completions", base_url=self._client_wrapper.get_environment().base, method="POST", json={ "messages": convert_and_respect_annotation_metadata( object_=messages, annotation=typing.Sequence[ChatCompletionRequestMessageParams], direction="write" ), "temperature": temperature, "top_p": top_p, "reasoning_effort": reasoning_effort, "max_tokens": max_tokens, "stream": stream, "stop": convert_and_respect_annotation_metadata( object_=stop, annotation=StopConfigurationParams, direction="write" ), "n": n, "seed": seed, "frequency_penalty": frequency_penalty, "presence_penalty": presence_penalty, "wiki_grounding": wiki_grounding, "model": "sarvam-m", }, headers={ "content-type": "application/json", }, request_options=request_options, omit=OMIT, ) try: if 200 <= _response.status_code < 300: _data = typing.cast( CreateChatCompletionResponse, parse_obj_as( type_=CreateChatCompletionResponse, # type: ignore object_=_response.json(), ), ) return HttpResponse(response=_response, data=_data) if _response.status_code == 400: raise BadRequestError( headers=dict(_response.headers), body=typing.cast( typing.Optional[typing.Any], parse_obj_as( type_=typing.Optional[typing.Any], # type: ignore object_=_response.json(), ), ), ) if _response.status_code == 403: raise ForbiddenError( headers=dict(_response.headers), body=typing.cast( typing.Optional[typing.Any], parse_obj_as( type_=typing.Optional[typing.Any], # type: ignore object_=_response.json(), ), ), ) if _response.status_code == 422: raise UnprocessableEntityError( headers=dict(_response.headers), body=typing.cast( typing.Optional[typing.Any], parse_obj_as( type_=typing.Optional[typing.Any], # type: ignore object_=_response.json(), ), ), ) if _response.status_code == 429: raise TooManyRequestsError( headers=dict(_response.headers), body=typing.cast( typing.Optional[typing.Any], parse_obj_as( type_=typing.Optional[typing.Any], # type: ignore object_=_response.json(), ), ), ) if _response.status_code == 500: raise InternalServerError( headers=dict(_response.headers), body=typing.cast( typing.Optional[typing.Any], parse_obj_as( type_=typing.Optional[typing.Any], # type: ignore object_=_response.json(), ), ), ) _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, headers=dict(_response.headers), body=_response.text) raise ApiError(status_code=_response.status_code, headers=dict(_response.headers), body=_response_json) class AsyncRawChatClient: def __init__(self, *, client_wrapper: AsyncClientWrapper): self._client_wrapper = client_wrapper async def completions( self, *, messages: typing.Sequence[ChatCompletionRequestMessageParams], temperature: typing.Optional[float] = OMIT, top_p: typing.Optional[float] = OMIT, reasoning_effort: typing.Optional[ReasoningEffort] = OMIT, max_tokens: typing.Optional[int] = OMIT, stream: typing.Optional[bool] = OMIT, stop: typing.Optional[StopConfigurationParams] = OMIT, n: typing.Optional[int] = OMIT, seed: typing.Optional[int] = OMIT, frequency_penalty: typing.Optional[float] = OMIT, presence_penalty: typing.Optional[float] = OMIT, wiki_grounding: typing.Optional[bool] = OMIT, request_options: typing.Optional[RequestOptions] = None, ) -> AsyncHttpResponse[CreateChatCompletionResponse]: """ Calls Sarvam LLM API to get the chat completion. Supported model(s): `sarvam-m`. Parameters ---------- messages : typing.Sequence[ChatCompletionRequestMessageParams] A list of messages comprising the conversation so far. temperature : typing.Optional[float] What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or `top_p` but not both. top_p : typing.Optional[float] An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered. We generally recommend altering this or `temperature` but not both. reasoning_effort : typing.Optional[ReasoningEffort] max_tokens : typing.Optional[int] The maximum number of tokens that can be generated in the chat completion. stream : typing.Optional[bool] If set to true, the model response data will be streamed to the client as it is generated using [server-sent events](https://developer.mozilla.org/en-US/docs/Web/API/Server-sent_events/Using_server-sent_events#Event_stream_format). stop : typing.Optional[StopConfigurationParams] n : typing.Optional[int] How many chat completion choices to generate for each input message. Note that you will be charged based on the number of generated tokens across all of the choices. Keep `n` as `1` to minimize costs. seed : typing.Optional[int] This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same `seed` and parameters should return the same result. frequency_penalty : typing.Optional[float] Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim. presence_penalty : typing.Optional[float] Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics. wiki_grounding : typing.Optional[bool] If this parameter is enabled, then the model uses a RAG based approach to retrieve relevant chunks from Wikipedia and uses them to answer the question. This is particularly useful for queries seeking factual information. request_options : typing.Optional[RequestOptions] Request-specific configuration. Returns ------- AsyncHttpResponse[CreateChatCompletionResponse] Successful Response """ _response = await self._client_wrapper.httpx_client.request( "v1/chat/completions", base_url=self._client_wrapper.get_environment().base, method="POST", json={ "messages": convert_and_respect_annotation_metadata( object_=messages, annotation=typing.Sequence[ChatCompletionRequestMessageParams], direction="write" ), "temperature": temperature, "top_p": top_p, "reasoning_effort": reasoning_effort, "max_tokens": max_tokens, "stream": stream, "stop": convert_and_respect_annotation_metadata( object_=stop, annotation=StopConfigurationParams, direction="write" ), "n": n, "seed": seed, "frequency_penalty": frequency_penalty, "presence_penalty": presence_penalty, "wiki_grounding": wiki_grounding, "model": "sarvam-m", }, headers={ "content-type": "application/json", }, request_options=request_options, omit=OMIT, ) try: if 200 <= _response.status_code < 300: _data = typing.cast( CreateChatCompletionResponse, parse_obj_as( type_=CreateChatCompletionResponse, # type: ignore object_=_response.json(), ), ) return AsyncHttpResponse(response=_response, data=_data) if _response.status_code == 400: raise BadRequestError( headers=dict(_response.headers), body=typing.cast( typing.Optional[typing.Any], parse_obj_as( type_=typing.Optional[typing.Any], # type: ignore object_=_response.json(), ), ), ) if _response.status_code == 403: raise ForbiddenError( headers=dict(_response.headers), body=typing.cast( typing.Optional[typing.Any], parse_obj_as( type_=typing.Optional[typing.Any], # type: ignore object_=_response.json(), ), ), ) if _response.status_code == 422: raise UnprocessableEntityError( headers=dict(_response.headers), body=typing.cast( typing.Optional[typing.Any], parse_obj_as( type_=typing.Optional[typing.Any], # type: ignore object_=_response.json(), ), ), ) if _response.status_code == 429: raise TooManyRequestsError( headers=dict(_response.headers), body=typing.cast( typing.Optional[typing.Any], parse_obj_as( type_=typing.Optional[typing.Any], # type: ignore object_=_response.json(), ), ), ) if _response.status_code == 500: raise InternalServerError( headers=dict(_response.headers), body=typing.cast( typing.Optional[typing.Any], parse_obj_as( type_=typing.Optional[typing.Any], # type: ignore object_=_response.json(), ), ), ) _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, headers=dict(_response.headers), body=_response.text) raise ApiError(status_code=_response.status_code, headers=dict(_response.headers), body=_response_json)