mistralai 1.0.2__py3-none-any.whl → 1.1.0__py3-none-any.whl

mistralai/chat.py

@@ -7,12 +7,13 @@ from mistralai.types import Nullable, OptionalNullable, UNSET
 from mistralai.utils import eventstreaming, get_security_from_env
 from typing import Any, AsyncGenerator, Generator, List, Optional, Union
 
+
 class Chat(BaseSDK):
     r"""Chat Completion API."""
-    
-    
+
     def complete(
-        self, *,
+        self,
+        *,
         model: Nullable[str],
         messages: Union[List[models.Messages], List[models.MessagesTypedDict]],
         temperature: Optional[float] = 0.7,
@@ -22,9 +23,18 @@ class Chat(BaseSDK):
         stream: Optional[bool] = False,
         stop: Optional[Union[models.Stop, models.StopTypedDict]] = None,
         random_seed: OptionalNullable[int] = UNSET,
-        response_format: Optional[Union[models.ResponseFormat, models.ResponseFormatTypedDict]] = None,
-        tools: OptionalNullable[Union[List[models.Tool], List[models.ToolTypedDict]]] = UNSET,
-        tool_choice: Optional[models.ToolChoice] = "auto",
+        response_format: Optional[
+            Union[models.ResponseFormat, models.ResponseFormatTypedDict]
+        ] = None,
+        tools: OptionalNullable[
+            Union[List[models.Tool], List[models.ToolTypedDict]]
+        ] = UNSET,
+        tool_choice: Optional[
+            Union[
+                models.ChatCompletionRequestToolChoice,
+                models.ChatCompletionRequestToolChoiceTypedDict,
+            ]
+        ] = None,
         safe_prompt: Optional[bool] = False,
         retries: OptionalNullable[utils.RetryConfig] = UNSET,
         server_url: Optional[str] = None,
@@ -41,9 +51,9 @@ class Chat(BaseSDK):
         :param stream: Whether to stream back partial progress. If set, tokens will be sent as data-only server-side events as they become available, with the stream terminated by a data: [DONE] message. Otherwise, the server will hold the request open until the timeout or until completion, with the response containing the full result as JSON.
         :param stop: Stop generation if this token is detected. Or if one of these tokens is detected when providing an array
         :param random_seed: The seed to use for random sampling. If set, different calls will generate deterministic results.
-        :param response_format: 
-        :param tools: 
-        :param tool_choice: 
+        :param response_format:
+        :param tools:
+        :param tool_choice:
         :param safe_prompt: Whether to inject a safety prompt before all conversations.
         :param retries: Override the default retry configuration for this method
         :param server_url: Override the default server URL for this method
@@ -53,10 +63,10 @@ class Chat(BaseSDK):
         url_variables = None
         if timeout_ms is None:
             timeout_ms = self.sdk_configuration.timeout_ms
-        
+
         if server_url is not None:
             base_url = server_url
-        
+
         request = models.ChatCompletionRequest(
             model=model,
             temperature=temperature,
@@ -67,12 +77,16 @@ class Chat(BaseSDK):
             stop=stop,
             random_seed=random_seed,
             messages=utils.get_pydantic_model(messages, List[models.Messages]),
-            response_format=utils.get_pydantic_model(response_format, Optional[models.ResponseFormat]),
+            response_format=utils.get_pydantic_model(
+                response_format, Optional[models.ResponseFormat]
+            ),
             tools=utils.get_pydantic_model(tools, OptionalNullable[List[models.Tool]]),
-            tool_choice=tool_choice,
+            tool_choice=utils.get_pydantic_model(
+                tool_choice, Optional[models.ChatCompletionRequestToolChoice]
+            ),
             safe_prompt=safe_prompt,
         )
-        
+
         req = self.build_request(
             method="POST",
             path="/v1/chat/completions",
@@ -85,47 +99,57 @@ class Chat(BaseSDK):
             user_agent_header="user-agent",
             accept_header_value="application/json",
             security=self.sdk_configuration.security,
-            get_serialized_body=lambda: utils.serialize_request_body(request, False, False, "json", models.ChatCompletionRequest),
+            get_serialized_body=lambda: utils.serialize_request_body(
+                request, False, False, "json", models.ChatCompletionRequest
+            ),
             timeout_ms=timeout_ms,
         )
-        
+
         if retries == UNSET:
             if self.sdk_configuration.retry_config is not UNSET:
                 retries = self.sdk_configuration.retry_config
 
         retry_config = None
         if isinstance(retries, utils.RetryConfig):
-            retry_config = (retries, [
-                "429",
-                "500",
-                "502",
-                "503",
-                "504"
-            ])                
-        
+            retry_config = (retries, ["429", "500", "502", "503", "504"])
+
         http_res = self.do_request(
-            hook_ctx=HookContext(operation_id="chat_completion_v1_chat_completions_post", oauth2_scopes=[], security_source=get_security_from_env(self.sdk_configuration.security, models.Security)),
+            hook_ctx=HookContext(
+                operation_id="chat_completion_v1_chat_completions_post",
+                oauth2_scopes=[],
+                security_source=get_security_from_env(
+                    self.sdk_configuration.security, models.Security
+                ),
+            ),
             request=req,
-            error_status_codes=["422","4XX","5XX"],
-            retry_config=retry_config
+            error_status_codes=["422", "4XX", "5XX"],
+            retry_config=retry_config,
         )
-        
+
         data: Any = None
         if utils.match_response(http_res, "200", "application/json"):
-            return utils.unmarshal_json(http_res.text, Optional[models.ChatCompletionResponse])
+            return utils.unmarshal_json(
+                http_res.text, Optional[models.ChatCompletionResponse]
+            )
         if utils.match_response(http_res, "422", "application/json"):
             data = utils.unmarshal_json(http_res.text, models.HTTPValidationErrorData)
             raise models.HTTPValidationError(data=data)
-        if utils.match_response(http_res, ["4XX","5XX"], "*"):
-            raise models.SDKError("API error occurred", http_res.status_code, http_res.text, http_res)
-        
+        if utils.match_response(http_res, ["4XX", "5XX"], "*"):
+            raise models.SDKError(
+                "API error occurred", http_res.status_code, http_res.text, http_res
+            )
+
         content_type = http_res.headers.get("Content-Type")
-        raise models.SDKError(f"Unexpected response received (code: {http_res.status_code}, type: {content_type})", http_res.status_code, http_res.text, http_res)
+        raise models.SDKError(
+            f"Unexpected response received (code: {http_res.status_code}, type: {content_type})",
+            http_res.status_code,
+            http_res.text,
+            http_res,
+        )
 
-    
-    
     async def complete_async(
-        self, *,
+        self,
+        *,
         model: Nullable[str],
         messages: Union[List[models.Messages], List[models.MessagesTypedDict]],
         temperature: Optional[float] = 0.7,
@@ -135,9 +159,18 @@ class Chat(BaseSDK):
         stream: Optional[bool] = False,
         stop: Optional[Union[models.Stop, models.StopTypedDict]] = None,
         random_seed: OptionalNullable[int] = UNSET,
-        response_format: Optional[Union[models.ResponseFormat, models.ResponseFormatTypedDict]] = None,
-        tools: OptionalNullable[Union[List[models.Tool], List[models.ToolTypedDict]]] = UNSET,
-        tool_choice: Optional[models.ToolChoice] = "auto",
+        response_format: Optional[
+            Union[models.ResponseFormat, models.ResponseFormatTypedDict]
+        ] = None,
+        tools: OptionalNullable[
+            Union[List[models.Tool], List[models.ToolTypedDict]]
+        ] = UNSET,
+        tool_choice: Optional[
+            Union[
+                models.ChatCompletionRequestToolChoice,
+                models.ChatCompletionRequestToolChoiceTypedDict,
+            ]
+        ] = None,
         safe_prompt: Optional[bool] = False,
         retries: OptionalNullable[utils.RetryConfig] = UNSET,
         server_url: Optional[str] = None,
@@ -154,9 +187,9 @@ class Chat(BaseSDK):
         :param stream: Whether to stream back partial progress. If set, tokens will be sent as data-only server-side events as they become available, with the stream terminated by a data: [DONE] message. Otherwise, the server will hold the request open until the timeout or until completion, with the response containing the full result as JSON.
         :param stop: Stop generation if this token is detected. Or if one of these tokens is detected when providing an array
         :param random_seed: The seed to use for random sampling. If set, different calls will generate deterministic results.
-        :param response_format: 
-        :param tools: 
-        :param tool_choice: 
+        :param response_format:
+        :param tools:
+        :param tool_choice:
         :param safe_prompt: Whether to inject a safety prompt before all conversations.
         :param retries: Override the default retry configuration for this method
         :param server_url: Override the default server URL for this method
@@ -166,10 +199,10 @@ class Chat(BaseSDK):
         url_variables = None
         if timeout_ms is None:
             timeout_ms = self.sdk_configuration.timeout_ms
-        
+
         if server_url is not None:
             base_url = server_url
-        
+
         request = models.ChatCompletionRequest(
             model=model,
             temperature=temperature,
@@ -180,13 +213,17 @@ class Chat(BaseSDK):
             stop=stop,
             random_seed=random_seed,
             messages=utils.get_pydantic_model(messages, List[models.Messages]),
-            response_format=utils.get_pydantic_model(response_format, Optional[models.ResponseFormat]),
+            response_format=utils.get_pydantic_model(
+                response_format, Optional[models.ResponseFormat]
+            ),
             tools=utils.get_pydantic_model(tools, OptionalNullable[List[models.Tool]]),
-            tool_choice=tool_choice,
+            tool_choice=utils.get_pydantic_model(
+                tool_choice, Optional[models.ChatCompletionRequestToolChoice]
+            ),
             safe_prompt=safe_prompt,
         )
-        
-        req = self.build_request(
+
+        req = self.build_request_async(
             method="POST",
             path="/v1/chat/completions",
             base_url=base_url,
@@ -198,59 +235,86 @@ class Chat(BaseSDK):
             user_agent_header="user-agent",
             accept_header_value="application/json",
             security=self.sdk_configuration.security,
-            get_serialized_body=lambda: utils.serialize_request_body(request, False, False, "json", models.ChatCompletionRequest),
+            get_serialized_body=lambda: utils.serialize_request_body(
+                request, False, False, "json", models.ChatCompletionRequest
+            ),
             timeout_ms=timeout_ms,
         )
-        
+
         if retries == UNSET:
             if self.sdk_configuration.retry_config is not UNSET:
                 retries = self.sdk_configuration.retry_config
 
         retry_config = None
         if isinstance(retries, utils.RetryConfig):
-            retry_config = (retries, [
-                "429",
-                "500",
-                "502",
-                "503",
-                "504"
-            ])                
-        
+            retry_config = (retries, ["429", "500", "502", "503", "504"])
+
         http_res = await self.do_request_async(
-            hook_ctx=HookContext(operation_id="chat_completion_v1_chat_completions_post", oauth2_scopes=[], security_source=get_security_from_env(self.sdk_configuration.security, models.Security)),
+            hook_ctx=HookContext(
+                operation_id="chat_completion_v1_chat_completions_post",
+                oauth2_scopes=[],
+                security_source=get_security_from_env(
+                    self.sdk_configuration.security, models.Security
+                ),
+            ),
             request=req,
-            error_status_codes=["422","4XX","5XX"],
-            retry_config=retry_config
+            error_status_codes=["422", "4XX", "5XX"],
+            retry_config=retry_config,
         )
-        
+
         data: Any = None
         if utils.match_response(http_res, "200", "application/json"):
-            return utils.unmarshal_json(http_res.text, Optional[models.ChatCompletionResponse])
+            return utils.unmarshal_json(
+                http_res.text, Optional[models.ChatCompletionResponse]
+            )
         if utils.match_response(http_res, "422", "application/json"):
             data = utils.unmarshal_json(http_res.text, models.HTTPValidationErrorData)
             raise models.HTTPValidationError(data=data)
-        if utils.match_response(http_res, ["4XX","5XX"], "*"):
-            raise models.SDKError("API error occurred", http_res.status_code, http_res.text, http_res)
-        
+        if utils.match_response(http_res, ["4XX", "5XX"], "*"):
+            raise models.SDKError(
+                "API error occurred", http_res.status_code, http_res.text, http_res
+            )
+
         content_type = http_res.headers.get("Content-Type")
-        raise models.SDKError(f"Unexpected response received (code: {http_res.status_code}, type: {content_type})", http_res.status_code, http_res.text, http_res)
+        raise models.SDKError(
+            f"Unexpected response received (code: {http_res.status_code}, type: {content_type})",
+            http_res.status_code,
+            http_res.text,
+            http_res,
+        )
 
-    
-    
     def stream(
-        self, *,
+        self,
+        *,
         model: Nullable[str],
-        messages: Union[List[models.ChatCompletionStreamRequestMessages], List[models.ChatCompletionStreamRequestMessagesTypedDict]],
+        messages: Union[
+            List[models.ChatCompletionStreamRequestMessages],
+            List[models.ChatCompletionStreamRequestMessagesTypedDict],
+        ],
         temperature: Optional[float] = 0.7,
         top_p: Optional[float] = 1,
         max_tokens: OptionalNullable[int] = UNSET,
         min_tokens: OptionalNullable[int] = UNSET,
         stream: Optional[bool] = True,
-        stop: Optional[Union[models.ChatCompletionStreamRequestStop, models.ChatCompletionStreamRequestStopTypedDict]] = None,
+        stop: Optional[
+            Union[
+                models.ChatCompletionStreamRequestStop,
+                models.ChatCompletionStreamRequestStopTypedDict,
+            ]
+        ] = None,
         random_seed: OptionalNullable[int] = UNSET,
-        response_format: Optional[Union[models.ResponseFormat, models.ResponseFormatTypedDict]] = None,
-        tools: OptionalNullable[Union[List[models.Tool], List[models.ToolTypedDict]]] = UNSET,
-        tool_choice: Optional[models.ChatCompletionStreamRequestToolChoice] = "auto",
+        response_format: Optional[
+            Union[models.ResponseFormat, models.ResponseFormatTypedDict]
+        ] = None,
+        tools: OptionalNullable[
+            Union[List[models.Tool], List[models.ToolTypedDict]]
+        ] = UNSET,
+        tool_choice: Optional[
+            Union[
+                models.ChatCompletionStreamRequestToolChoice,
+                models.ChatCompletionStreamRequestToolChoiceTypedDict,
+            ]
+        ] = None,
         safe_prompt: Optional[bool] = False,
         retries: OptionalNullable[utils.RetryConfig] = UNSET,
         server_url: Optional[str] = None,
@@ -266,12 +330,12 @@ class Chat(BaseSDK):
         :param top_p: 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.
         :param max_tokens: The maximum number of tokens to generate in the completion. The token count of your prompt plus `max_tokens` cannot exceed the model's context length.
         :param min_tokens: The minimum number of tokens to generate in the completion.
-        :param stream: 
+        :param stream:
         :param stop: Stop generation if this token is detected. Or if one of these tokens is detected when providing an array
         :param random_seed: The seed to use for random sampling. If set, different calls will generate deterministic results.
-        :param response_format: 
-        :param tools: 
-        :param tool_choice: 
+        :param response_format:
+        :param tools:
+        :param tool_choice:
         :param safe_prompt: Whether to inject a safety prompt before all conversations.
         :param retries: Override the default retry configuration for this method
         :param server_url: Override the default server URL for this method
@@ -281,10 +345,10 @@ class Chat(BaseSDK):
         url_variables = None
         if timeout_ms is None:
             timeout_ms = self.sdk_configuration.timeout_ms
-        
+
         if server_url is not None:
             base_url = server_url
-        
+
         request = models.ChatCompletionStreamRequest(
             model=model,
             temperature=temperature,
@@ -294,13 +358,19 @@ class Chat(BaseSDK):
             stream=stream,
             stop=stop,
             random_seed=random_seed,
-            messages=utils.get_pydantic_model(messages, List[models.ChatCompletionStreamRequestMessages]),
-            response_format=utils.get_pydantic_model(response_format, Optional[models.ResponseFormat]),
+            messages=utils.get_pydantic_model(
+                messages, List[models.ChatCompletionStreamRequestMessages]
+            ),
+            response_format=utils.get_pydantic_model(
+                response_format, Optional[models.ResponseFormat]
+            ),
             tools=utils.get_pydantic_model(tools, OptionalNullable[List[models.Tool]]),
-            tool_choice=tool_choice,
+            tool_choice=utils.get_pydantic_model(
+                tool_choice, Optional[models.ChatCompletionStreamRequestToolChoice]
+            ),
             safe_prompt=safe_prompt,
         )
-        
+
         req = self.build_request(
             method="POST",
             path="/v1/chat/completions#stream",
@@ -313,60 +383,89 @@ class Chat(BaseSDK):
             user_agent_header="user-agent",
             accept_header_value="text/event-stream",
             security=self.sdk_configuration.security,
-            get_serialized_body=lambda: utils.serialize_request_body(request, False, False, "json", models.ChatCompletionStreamRequest),
+            get_serialized_body=lambda: utils.serialize_request_body(
+                request, False, False, "json", models.ChatCompletionStreamRequest
+            ),
             timeout_ms=timeout_ms,
         )
-        
+
         if retries == UNSET:
             if self.sdk_configuration.retry_config is not UNSET:
                 retries = self.sdk_configuration.retry_config
 
         retry_config = None
         if isinstance(retries, utils.RetryConfig):
-            retry_config = (retries, [
-                "429",
-                "500",
-                "502",
-                "503",
-                "504"
-            ])                
-        
+            retry_config = (retries, ["429", "500", "502", "503", "504"])
+
         http_res = self.do_request(
-            hook_ctx=HookContext(operation_id="stream_chat", oauth2_scopes=[], security_source=get_security_from_env(self.sdk_configuration.security, models.Security)),
+            hook_ctx=HookContext(
+                operation_id="stream_chat",
+                oauth2_scopes=[],
+                security_source=get_security_from_env(
+                    self.sdk_configuration.security, models.Security
+                ),
+            ),
             request=req,
-            error_status_codes=["422","4XX","5XX"],
+            error_status_codes=["422", "4XX", "5XX"],
             stream=True,
-            retry_config=retry_config
+            retry_config=retry_config,
         )
-        
+
         data: Any = None
         if utils.match_response(http_res, "200", "text/event-stream"):
-            return eventstreaming.stream_events(http_res, lambda raw: utils.unmarshal_json(raw, models.CompletionEvent), sentinel="[DONE]")
+            return eventstreaming.stream_events(
+                http_res,
+                lambda raw: utils.unmarshal_json(raw, models.CompletionEvent),
+                sentinel="[DONE]",
+            )
         if utils.match_response(http_res, "422", "application/json"):
             data = utils.unmarshal_json(http_res.text, models.HTTPValidationErrorData)
             raise models.HTTPValidationError(data=data)
-        if utils.match_response(http_res, ["4XX","5XX"], "*"):
-            raise models.SDKError("API error occurred", http_res.status_code, http_res.text, http_res)
-        
+        if utils.match_response(http_res, ["4XX", "5XX"], "*"):
+            raise models.SDKError(
+                "API error occurred", http_res.status_code, http_res.text, http_res
+            )
+
         content_type = http_res.headers.get("Content-Type")
-        raise models.SDKError(f"Unexpected response received (code: {http_res.status_code}, type: {content_type})", http_res.status_code, http_res.text, http_res)
+        raise models.SDKError(
+            f"Unexpected response received (code: {http_res.status_code}, type: {content_type})",
+            http_res.status_code,
+            http_res.text,
+            http_res,
+        )
 
-    
-    
     async def stream_async(
-        self, *,
+        self,
+        *,
         model: Nullable[str],
-        messages: Union[List[models.ChatCompletionStreamRequestMessages], List[models.ChatCompletionStreamRequestMessagesTypedDict]],
+        messages: Union[
+            List[models.ChatCompletionStreamRequestMessages],
+            List[models.ChatCompletionStreamRequestMessagesTypedDict],
+        ],
         temperature: Optional[float] = 0.7,
         top_p: Optional[float] = 1,
         max_tokens: OptionalNullable[int] = UNSET,
         min_tokens: OptionalNullable[int] = UNSET,
         stream: Optional[bool] = True,
-        stop: Optional[Union[models.ChatCompletionStreamRequestStop, models.ChatCompletionStreamRequestStopTypedDict]] = None,
+        stop: Optional[
+            Union[
+                models.ChatCompletionStreamRequestStop,
+                models.ChatCompletionStreamRequestStopTypedDict,
+            ]
+        ] = None,
         random_seed: OptionalNullable[int] = UNSET,
-        response_format: Optional[Union[models.ResponseFormat, models.ResponseFormatTypedDict]] = None,
-        tools: OptionalNullable[Union[List[models.Tool], List[models.ToolTypedDict]]] = UNSET,
-        tool_choice: Optional[models.ChatCompletionStreamRequestToolChoice] = "auto",
+        response_format: Optional[
+            Union[models.ResponseFormat, models.ResponseFormatTypedDict]
+        ] = None,
+        tools: OptionalNullable[
+            Union[List[models.Tool], List[models.ToolTypedDict]]
+        ] = UNSET,
+        tool_choice: Optional[
+            Union[
+                models.ChatCompletionStreamRequestToolChoice,
+                models.ChatCompletionStreamRequestToolChoiceTypedDict,
+            ]
+        ] = None,
         safe_prompt: Optional[bool] = False,
         retries: OptionalNullable[utils.RetryConfig] = UNSET,
         server_url: Optional[str] = None,
@@ -382,12 +481,12 @@ class Chat(BaseSDK):
         :param top_p: 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.
         :param max_tokens: The maximum number of tokens to generate in the completion. The token count of your prompt plus `max_tokens` cannot exceed the model's context length.
         :param min_tokens: The minimum number of tokens to generate in the completion.
-        :param stream: 
+        :param stream:
         :param stop: Stop generation if this token is detected. Or if one of these tokens is detected when providing an array
         :param random_seed: The seed to use for random sampling. If set, different calls will generate deterministic results.
-        :param response_format: 
-        :param tools: 
-        :param tool_choice: 
+        :param response_format:
+        :param tools:
+        :param tool_choice:
         :param safe_prompt: Whether to inject a safety prompt before all conversations.
         :param retries: Override the default retry configuration for this method
         :param server_url: Override the default server URL for this method
@@ -397,10 +496,10 @@ class Chat(BaseSDK):
         url_variables = None
         if timeout_ms is None:
             timeout_ms = self.sdk_configuration.timeout_ms
-        
+
         if server_url is not None:
             base_url = server_url
-        
+
         request = models.ChatCompletionStreamRequest(
             model=model,
             temperature=temperature,
@@ -410,14 +509,20 @@ class Chat(BaseSDK):
             stream=stream,
             stop=stop,
             random_seed=random_seed,
-            messages=utils.get_pydantic_model(messages, List[models.ChatCompletionStreamRequestMessages]),
-            response_format=utils.get_pydantic_model(response_format, Optional[models.ResponseFormat]),
+            messages=utils.get_pydantic_model(
+                messages, List[models.ChatCompletionStreamRequestMessages]
+            ),
+            response_format=utils.get_pydantic_model(
+                response_format, Optional[models.ResponseFormat]
+            ),
             tools=utils.get_pydantic_model(tools, OptionalNullable[List[models.Tool]]),
-            tool_choice=tool_choice,
+            tool_choice=utils.get_pydantic_model(
+                tool_choice, Optional[models.ChatCompletionStreamRequestToolChoice]
+            ),
             safe_prompt=safe_prompt,
         )
-        
-        req = self.build_request(
+
+        req = self.build_request_async(
             method="POST",
             path="/v1/chat/completions#stream",
             base_url=base_url,
@@ -429,42 +534,53 @@ class Chat(BaseSDK):
             user_agent_header="user-agent",
             accept_header_value="text/event-stream",
             security=self.sdk_configuration.security,
-            get_serialized_body=lambda: utils.serialize_request_body(request, False, False, "json", models.ChatCompletionStreamRequest),
+            get_serialized_body=lambda: utils.serialize_request_body(
+                request, False, False, "json", models.ChatCompletionStreamRequest
+            ),
             timeout_ms=timeout_ms,
         )
-        
+
         if retries == UNSET:
             if self.sdk_configuration.retry_config is not UNSET:
                 retries = self.sdk_configuration.retry_config
 
         retry_config = None
         if isinstance(retries, utils.RetryConfig):
-            retry_config = (retries, [
-                "429",
-                "500",
-                "502",
-                "503",
-                "504"
-            ])                
-        
+            retry_config = (retries, ["429", "500", "502", "503", "504"])
+
         http_res = await self.do_request_async(
-            hook_ctx=HookContext(operation_id="stream_chat", oauth2_scopes=[], security_source=get_security_from_env(self.sdk_configuration.security, models.Security)),
+            hook_ctx=HookContext(
+                operation_id="stream_chat",
+                oauth2_scopes=[],
+                security_source=get_security_from_env(
+                    self.sdk_configuration.security, models.Security
+                ),
+            ),
             request=req,
-            error_status_codes=["422","4XX","5XX"],
+            error_status_codes=["422", "4XX", "5XX"],
             stream=True,
-            retry_config=retry_config
+            retry_config=retry_config,
         )
-        
+
         data: Any = None
         if utils.match_response(http_res, "200", "text/event-stream"):
-            return eventstreaming.stream_events_async(http_res, lambda raw: utils.unmarshal_json(raw, models.CompletionEvent), sentinel="[DONE]")
+            return eventstreaming.stream_events_async(
+                http_res,
+                lambda raw: utils.unmarshal_json(raw, models.CompletionEvent),
+                sentinel="[DONE]",
+            )
         if utils.match_response(http_res, "422", "application/json"):
             data = utils.unmarshal_json(http_res.text, models.HTTPValidationErrorData)
             raise models.HTTPValidationError(data=data)
-        if utils.match_response(http_res, ["4XX","5XX"], "*"):
-            raise models.SDKError("API error occurred", http_res.status_code, http_res.text, http_res)
-        
-        content_type = http_res.headers.get("Content-Type")
-        raise models.SDKError(f"Unexpected response received (code: {http_res.status_code}, type: {content_type})", http_res.status_code, http_res.text, http_res)
+        if utils.match_response(http_res, ["4XX", "5XX"], "*"):
+            raise models.SDKError(
+                "API error occurred", http_res.status_code, http_res.text, http_res
+            )
 
-    
+        content_type = http_res.headers.get("Content-Type")
+        raise models.SDKError(
+            f"Unexpected response received (code: {http_res.status_code}, type: {content_type})",
+            http_res.status_code,
+            http_res.text,
+            http_res,
+        )