[Feat] Add support for GET Responses Endpoint - OpenAI, Azure OpenAI (#10235)

* Added get responses API (#10234)

* test_basic_openai_responses_get_endpoint

* transform_get_response_api_request

* test_basic_openai_responses_get_endpoint

---------

Co-authored-by: Prathamesh Saraf <pratamesh1867@gmail.com>

litellm/llms/azure/responses/transformation.py

@@ -95,6 +95,35 @@ class AzureOpenAIResponsesAPIConfig(OpenAIResponsesAPIConfig):
     #########################################################
     ########## DELETE RESPONSE API TRANSFORMATION ##############
     #########################################################
+    def _construct_url_for_response_id_in_path(
+        self, api_base: str, response_id: str
+    ) -> str:
+        """
+        Constructs a URL for the API request with the response_id in the path.
+        """
+        from urllib.parse import urlparse, urlunparse
+
+        # Parse the URL to separate its components
+        parsed_url = urlparse(api_base)
+
+        # Insert the response_id at the end of the path component
+        # Remove trailing slash if present to avoid double slashes
+        path = parsed_url.path.rstrip("/")
+        new_path = f"{path}/{response_id}"
+
+        # Reconstruct the URL with all original components but with the modified path
+        constructed_url = urlunparse(
+            (
+                parsed_url.scheme,  # http, https
+                parsed_url.netloc,  # domain name, port
+                new_path,  # path with response_id added
+                parsed_url.params,  # parameters
+                parsed_url.query,  # query string
+                parsed_url.fragment,  # fragment
+            )
+        )
+        return constructed_url
+
     def transform_delete_response_api_request(
         self,
         response_id: str,
@@ -111,28 +140,33 @@ class AzureOpenAIResponsesAPIConfig(OpenAIResponsesAPIConfig):
         This function handles URLs with query parameters by inserting the response_id
         at the correct location (before any query parameters).
         """
-        from urllib.parse import urlparse, urlunparse
+        delete_url = self._construct_url_for_response_id_in_path(
-
+            api_base=api_base, response_id=response_id
-        # Parse the URL to separate its components
-        parsed_url = urlparse(api_base)
-
-        # Insert the response_id at the end of the path component
-        # Remove trailing slash if present to avoid double slashes
-        path = parsed_url.path.rstrip("/")
-        new_path = f"{path}/{response_id}"
-
-        # Reconstruct the URL with all original components but with the modified path
-        delete_url = urlunparse(
-            (
-                parsed_url.scheme,  # http, https
-                parsed_url.netloc,  # domain name, port
-                new_path,  # path with response_id added
-                parsed_url.params,  # parameters
-                parsed_url.query,  # query string
-                parsed_url.fragment,  # fragment
-            )
         )
 
         data: Dict = {}
         verbose_logger.debug(f"delete response url={delete_url}")
         return delete_url, data
+
+    #########################################################
+    ########## GET RESPONSE API TRANSFORMATION ###############
+    #########################################################
+    def transform_get_response_api_request(
+        self,
+        response_id: str,
+        api_base: str,
+        litellm_params: GenericLiteLLMParams,
+        headers: dict,
+    ) -> Tuple[str, Dict]:
+        """
+        Transform the get response API request into a URL and data
+
+        OpenAI API expects the following request
+        - GET /v1/responses/{response_id}
+        """
+        get_url = self._construct_url_for_response_id_in_path(
+            api_base=api_base, response_id=response_id
+        )
+        data: Dict = {}
+        verbose_logger.debug(f"get response url={get_url}")
+        return get_url, data

litellm/llms/base_llm/responses/transformation.py

@@ -141,9 +141,34 @@ class BaseResponsesAPIConfig(ABC):
         pass
 
     #########################################################
-    ########## END DELETE RESPONSE API TRANSFORMATION ##########
+    ########## END DELETE RESPONSE API TRANSFORMATION #######
     #########################################################
 
+    #########################################################
+    ########## GET RESPONSE API TRANSFORMATION ###############
+    #########################################################
+    @abstractmethod
+    def transform_get_response_api_request(
+        self,
+        response_id: str,
+        api_base: str,
+        litellm_params: GenericLiteLLMParams,
+        headers: dict,
+    ) -> Tuple[str, Dict]:
+        pass
+    
+    @abstractmethod
+    def transform_get_response_api_response(
+        self,
+        raw_response: httpx.Response,
+        logging_obj: LiteLLMLoggingObj,
+    ) -> ResponsesAPIResponse:
+        pass
+
+    #########################################################
+    ########## END GET RESPONSE API TRANSFORMATION ##########
+    #########################################################
+    
     def get_error_class(
         self, error_message: str, status_code: int, headers: Union[dict, httpx.Headers]
     ) -> BaseLLMException:

litellm/llms/custom_httpx/llm_http_handler.py

@@ -1426,6 +1426,162 @@ class BaseLLMHTTPHandler:
             logging_obj=logging_obj,
         )
 
+    def get_responses(
+        self,
+        response_id: str,
+        responses_api_provider_config: BaseResponsesAPIConfig,
+        litellm_params: GenericLiteLLMParams,
+        logging_obj: LiteLLMLoggingObj,
+        custom_llm_provider: Optional[str] = None,
+        extra_headers: Optional[Dict[str, Any]] = None,
+        extra_body: Optional[Dict[str, Any]] = None,
+        timeout: Optional[Union[float, httpx.Timeout]] = None,
+        client: Optional[Union[HTTPHandler, AsyncHTTPHandler]] = None,
+        _is_async: bool = False,
+    ) -> Union[ResponsesAPIResponse, Coroutine[Any, Any, ResponsesAPIResponse]]:
+        """
+        Get a response by ID
+        Uses GET /v1/responses/{response_id} endpoint in the responses API
+        """
+        if _is_async:
+            return self.async_get_responses(
+                response_id=response_id,
+                responses_api_provider_config=responses_api_provider_config,
+                litellm_params=litellm_params,
+                logging_obj=logging_obj,
+                custom_llm_provider=custom_llm_provider,
+                extra_headers=extra_headers,
+                extra_body=extra_body,
+                timeout=timeout,
+                client=client,
+            )
+        
+        if client is None or not isinstance(client, HTTPHandler):
+            sync_httpx_client = _get_httpx_client(
+                params={"ssl_verify": litellm_params.get("ssl_verify", None)}
+            )
+        else:
+            sync_httpx_client = client
+
+        headers = responses_api_provider_config.validate_environment(
+            api_key=litellm_params.api_key,
+            headers=extra_headers or {},
+            model="None",
+        )
+
+        if extra_headers:
+            headers.update(extra_headers)
+
+        api_base = responses_api_provider_config.get_complete_url(
+            api_base=litellm_params.api_base,
+            litellm_params=dict(litellm_params),
+        )
+
+        url, data = responses_api_provider_config.transform_get_response_api_request(
+            response_id=response_id,
+            api_base=api_base,
+            litellm_params=litellm_params,
+            headers=headers,
+        )
+
+        ## LOGGING
+        logging_obj.pre_call(
+            input="",
+            api_key="",
+            additional_args={
+                "complete_input_dict": data,
+                "api_base": api_base,
+                "headers": headers,
+            },
+        )
+
+        try:
+            response = sync_httpx_client.get(
+                url=url, headers=headers, params=data
+            )
+        except Exception as e:
+            raise self._handle_error(
+                e=e,
+                provider_config=responses_api_provider_config,
+            )
+
+        return responses_api_provider_config.transform_get_response_api_response(
+            raw_response=response,
+            logging_obj=logging_obj,
+        )
+
+    async def async_get_responses(
+        self,
+        response_id: str,
+        responses_api_provider_config: BaseResponsesAPIConfig,
+        litellm_params: GenericLiteLLMParams,
+        logging_obj: LiteLLMLoggingObj,
+        custom_llm_provider: Optional[str] = None,
+        extra_headers: Optional[Dict[str, Any]] = None,
+        extra_body: Optional[Dict[str, Any]] = None,
+        timeout: Optional[Union[float, httpx.Timeout]] = None,
+        client: Optional[Union[HTTPHandler, AsyncHTTPHandler]] = None,
+    ) -> ResponsesAPIResponse:
+        """
+        Async version of get_responses
+        """
+        if client is None or not isinstance(client, AsyncHTTPHandler):
+            async_httpx_client = get_async_httpx_client(
+                llm_provider=litellm.LlmProviders(custom_llm_provider),
+                params={"ssl_verify": litellm_params.get("ssl_verify", None)},
+            )
+        else:
+            async_httpx_client = client
+
+        headers = responses_api_provider_config.validate_environment(
+            api_key=litellm_params.api_key,
+            headers=extra_headers or {},
+            model="None",
+        )
+
+        if extra_headers:
+            headers.update(extra_headers)
+
+        api_base = responses_api_provider_config.get_complete_url(
+            api_base=litellm_params.api_base,
+            litellm_params=dict(litellm_params),
+        )
+
+        url, data = responses_api_provider_config.transform_get_response_api_request(
+            response_id=response_id,
+            api_base=api_base,
+            litellm_params=litellm_params,
+            headers=headers,
+        )
+
+        ## LOGGING
+        logging_obj.pre_call(
+            input="",
+            api_key="",
+            additional_args={
+                "complete_input_dict": data,
+                "api_base": api_base,
+                "headers": headers,
+            },
+        )
+
+        try:
+            response = await async_httpx_client.get(
+                url=url, headers=headers, params=data
+            )
+
+        except Exception as e:
+            verbose_logger.exception(f"Error retrieving response: {e}")
+            raise self._handle_error(
+                e=e,
+                provider_config=responses_api_provider_config,
+            )
+
+        return responses_api_provider_config.transform_get_response_api_response(
+            raw_response=response,
+            logging_obj=logging_obj,
+        )
+
     def create_file(
         self,
         create_file_data: CreateFileRequest,

litellm/llms/openai/responses/transformation.py

@@ -250,3 +250,39 @@ class OpenAIResponsesAPIConfig(BaseResponsesAPIConfig):
                 message=raw_response.text, status_code=raw_response.status_code
             )
         return DeleteResponseResult(**raw_response_json)
+    
+    #########################################################
+    ########## GET RESPONSE API TRANSFORMATION ###############
+    #########################################################
+    def transform_get_response_api_request(
+        self,
+        response_id: str,
+        api_base: str,
+        litellm_params: GenericLiteLLMParams,
+        headers: dict,
+    ) -> Tuple[str, Dict]:
+        """
+        Transform the get response API request into a URL and data
+
+        OpenAI API expects the following request
+        - GET /v1/responses/{response_id}
+        """
+        url = f"{api_base}/{response_id}"
+        data: Dict = {}
+        return url, data
+    
+    def transform_get_response_api_response(
+        self,
+        raw_response: httpx.Response,
+        logging_obj: LiteLLMLoggingObj,
+    ) -> ResponsesAPIResponse:
+        """
+        Transform the get response API response into a ResponsesAPIResponse
+        """
+        try:
+            raw_response_json = raw_response.json()
+        except Exception:
+            raise OpenAIError(
+                message=raw_response.text, status_code=raw_response.status_code
+            )
+        return ResponsesAPIResponse(**raw_response_json)

litellm/responses/main.py

@@ -434,3 +434,188 @@ def delete_responses(
             completion_kwargs=local_vars,
             extra_kwargs=kwargs,
         )
+
+@client
+async def aget_responses(
+    response_id: str,
+    # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+    # The extra values given here take precedence over values defined on the client or passed to this method.
+    extra_headers: Optional[Dict[str, Any]] = None,
+    extra_query: Optional[Dict[str, Any]] = None,
+    extra_body: Optional[Dict[str, Any]] = None,
+    timeout: Optional[Union[float, httpx.Timeout]] = None,
+    # LiteLLM specific params,
+    custom_llm_provider: Optional[str] = None,
+    **kwargs,
+) -> ResponsesAPIResponse:
+    """
+    Async: Fetch a response by its ID.
+    
+    GET /v1/responses/{response_id} endpoint in the responses API
+    
+    Args:
+        response_id: The ID of the response to fetch.
+        custom_llm_provider: Optional provider name. If not specified, will be decoded from response_id.
+        
+    Returns:
+        The response object with complete information about the stored response.
+    """
+    local_vars = locals()
+    try:
+        loop = asyncio.get_event_loop()
+        kwargs["aget_responses"] = True
+
+        # get custom llm provider from response_id
+        decoded_response_id: DecodedResponseId = (
+            ResponsesAPIRequestUtils._decode_responses_api_response_id(
+                response_id=response_id,
+            )
+        )
+        response_id = decoded_response_id.get("response_id") or response_id
+        custom_llm_provider = (
+            decoded_response_id.get("custom_llm_provider") or custom_llm_provider
+        )
+
+        func = partial(
+            get_responses,
+            response_id=response_id,
+            custom_llm_provider=custom_llm_provider,
+            extra_headers=extra_headers,
+            extra_query=extra_query,
+            extra_body=extra_body,
+            timeout=timeout,
+            **kwargs,
+        )
+
+        ctx = contextvars.copy_context()
+        func_with_context = partial(ctx.run, func)
+        init_response = await loop.run_in_executor(None, func_with_context)
+
+        if asyncio.iscoroutine(init_response):
+            response = await init_response
+        else:
+            response = init_response
+
+         # Update the responses_api_response_id with the model_id
+        if isinstance(response, ResponsesAPIResponse):
+            response = ResponsesAPIRequestUtils._update_responses_api_response_id_with_model_id(
+                responses_api_response=response,
+                litellm_metadata=kwargs.get("litellm_metadata", {}),
+                custom_llm_provider=custom_llm_provider,
+            )
+        return response
+    except Exception as e:
+        raise litellm.exception_type(
+            model=None,
+            custom_llm_provider=custom_llm_provider,
+            original_exception=e,
+            completion_kwargs=local_vars,
+            extra_kwargs=kwargs,
+        )
+
+@client
+def get_responses(
+    response_id: str,
+    # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+    # The extra values given here take precedence over values defined on the client or passed to this method.
+    extra_headers: Optional[Dict[str, Any]] = None,
+    extra_query: Optional[Dict[str, Any]] = None,
+    extra_body: Optional[Dict[str, Any]] = None,
+    timeout: Optional[Union[float, httpx.Timeout]] = None,
+    # LiteLLM specific params,
+    custom_llm_provider: Optional[str] = None,
+    **kwargs,
+) -> Union[ResponsesAPIResponse, Coroutine[Any, Any, ResponsesAPIResponse]]:
+    """
+    Fetch a response by its ID.
+    
+    GET /v1/responses/{response_id} endpoint in the responses API
+    
+    Args:
+        response_id: The ID of the response to fetch.
+        custom_llm_provider: Optional provider name. If not specified, will be decoded from response_id.
+        
+    Returns:
+        The response object with complete information about the stored response.
+    """
+    local_vars = locals()
+    try:
+        litellm_logging_obj: LiteLLMLoggingObj = kwargs.get("litellm_logging_obj")  # type: ignore
+        litellm_call_id: Optional[str] = kwargs.get("litellm_call_id", None)
+        _is_async = kwargs.pop("aget_responses", False) is True
+
+        # get llm provider logic
+        litellm_params = GenericLiteLLMParams(**kwargs)
+
+        # get custom llm provider from response_id
+        decoded_response_id: DecodedResponseId = (
+            ResponsesAPIRequestUtils._decode_responses_api_response_id(
+                response_id=response_id,
+            )
+        )
+        response_id = decoded_response_id.get("response_id") or response_id
+        custom_llm_provider = (
+            decoded_response_id.get("custom_llm_provider") or custom_llm_provider
+        )
+
+        if custom_llm_provider is None:
+            raise ValueError("custom_llm_provider is required but passed as None")
+
+        # get provider config
+        responses_api_provider_config: Optional[BaseResponsesAPIConfig] = (
+            ProviderConfigManager.get_provider_responses_api_config(
+                model=None,
+                provider=litellm.LlmProviders(custom_llm_provider),
+            )
+        )
+
+        if responses_api_provider_config is None:
+            raise ValueError(
+                f"GET responses is not supported for {custom_llm_provider}"
+            )
+
+        local_vars.update(kwargs)
+
+        # Pre Call logging
+        litellm_logging_obj.update_environment_variables(
+            model=None,
+            optional_params={
+                "response_id": response_id,
+            },
+            litellm_params={
+                "litellm_call_id": litellm_call_id,
+            },
+            custom_llm_provider=custom_llm_provider,
+        )
+
+        # Call the handler with _is_async flag instead of directly calling the async handler
+        response = base_llm_http_handler.get_responses(
+            response_id=response_id,
+            custom_llm_provider=custom_llm_provider,
+            responses_api_provider_config=responses_api_provider_config,
+            litellm_params=litellm_params,
+            logging_obj=litellm_logging_obj,
+            extra_headers=extra_headers,
+            extra_body=extra_body,
+            timeout=timeout or request_timeout,
+            _is_async=_is_async,
+            client=kwargs.get("client"),
+        )
+
+        # Update the responses_api_response_id with the model_id
+        if isinstance(response, ResponsesAPIResponse):
+            response = ResponsesAPIRequestUtils._update_responses_api_response_id_with_model_id(
+                responses_api_response=response,
+                litellm_metadata=kwargs.get("litellm_metadata", {}),
+                custom_llm_provider=custom_llm_provider,
+            )
+
+        return response
+    except Exception as e:
+        raise litellm.exception_type(
+            model=None,
+            custom_llm_provider=custom_llm_provider,
+            original_exception=e,
+            completion_kwargs=local_vars,
+            extra_kwargs=kwargs,
+        )

tests/llm_responses_api_testing/base_responses_api.py

@@ -274,5 +274,44 @@ class BaseResponsesAPITest(ABC):
                 **base_completion_call_args
             )
 
+    @pytest.mark.parametrize("sync_mode", [False, True])
+    @pytest.mark.asyncio
+    async def test_basic_openai_responses_get_endpoint(self, sync_mode):
+        litellm._turn_on_debug()
+        litellm.set_verbose = True
+        base_completion_call_args = self.get_base_completion_call_args()
+        if sync_mode:
+            response = litellm.responses(
+                input="Basic ping", max_output_tokens=20,
+                **base_completion_call_args
+            )
 
+            # get the response
+            if isinstance(response, ResponsesAPIResponse):
+                result = litellm.get_responses(
+                    response_id=response.id,
+                    **base_completion_call_args
+                )
+                assert result is not None
+                assert result.id == response.id
+                assert result.output == response.output
+            else:
+                raise ValueError("response is not a ResponsesAPIResponse")
+        else:
+            response = await litellm.aresponses(
+                input="Basic ping", max_output_tokens=20,
+                **base_completion_call_args
+            )
+            # async get the response
+            if isinstance(response, ResponsesAPIResponse):
+                result = await litellm.aget_responses(
+                    response_id=response.id,
+                    **base_completion_call_args
+                )
+                assert result is not None
+                assert result.id == response.id
+                assert result.output == response.output
+            else:
+                raise ValueError("response is not a ResponsesAPIResponse")
+    
 

tests/llm_responses_api_testing/test_anthropic_responses_api.py

@@ -36,6 +36,11 @@ class TestAnthropicResponsesAPITest(BaseResponsesAPITest):
     async def test_basic_openai_responses_streaming_delete_endpoint(self, sync_mode=False):
         pass
 
+    async def test_basic_openai_responses_get_endpoint(self, sync_mode=False):
+        pass
+    
+
+
 
 def test_multiturn_tool_calls():
     # Test streaming response with tools for Anthropic