mistralai 1.3.1__py3-none-any.whl → 1.5.0__py3-none-any.whl

mistralai/__init__.py

@@ -1,9 +1,18 @@
 """Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT."""
 
-from ._version import __title__, __version__
+from ._version import (
+    __title__,
+    __version__,
+    __openapi_doc_version__,
+    __gen_version__,
+    __user_agent__,
+)
 from .sdk import *
 from .sdkconfiguration import *
 from .models import *
 
 
 VERSION: str = __version__
+OPENAPI_DOC_VERSION = __openapi_doc_version__
+SPEAKEASY_GENERATOR_VERSION = __gen_version__
+USER_AGENT = __user_agent__

mistralai/_version.py

@@ -3,7 +3,10 @@
 import importlib.metadata
 
 __title__: str = "mistralai"
-__version__: str = "1.3.1"
+__version__: str = "1.5.0"
+__openapi_doc_version__: str = "0.0.2"
+__gen_version__: str = "2.497.0"
+__user_agent__: str = "speakeasy-sdk/python 1.5.0 2.497.0 0.0.2 mistralai"
 
 try:
     if __package__ is not None:

mistralai/agents.py

@@ -43,11 +43,14 @@ class Agents(BaseSDK):
         presence_penalty: Optional[float] = None,
         frequency_penalty: Optional[float] = None,
         n: OptionalNullable[int] = UNSET,
+        prediction: Optional[
+            Union[models.Prediction, models.PredictionTypedDict]
+        ] = None,
         retries: OptionalNullable[utils.RetryConfig] = UNSET,
         server_url: Optional[str] = None,
         timeout_ms: Optional[int] = None,
         http_headers: Optional[Mapping[str, str]] = None,
-    ) -> Optional[models.ChatCompletionResponse]:
+    ) -> models.ChatCompletionResponse:
         r"""Agents Completion
 
         :param messages: The prompt(s) to generate completions for, encoded as a list of dict with role and content.
@@ -62,6 +65,7 @@ class Agents(BaseSDK):
         :param presence_penalty: presence_penalty determines how much the model penalizes the repetition of words or phrases. A higher presence penalty encourages the model to use a wider variety of words and phrases, making the output more diverse and creative.
         :param frequency_penalty: frequency_penalty penalizes the repetition of words based on their frequency in the generated text. A higher frequency penalty discourages the model from repeating words that have already appeared frequently in the output, promoting diversity and reducing repetition.
         :param n: Number of completions to return for each request, input tokens are only billed once.
+        :param prediction:
         :param retries: Override the default retry configuration for this method
         :param server_url: Override the default server URL for this method
         :param timeout_ms: Override the default request timeout configuration for this method in milliseconds
@@ -93,6 +97,9 @@ class Agents(BaseSDK):
             presence_penalty=presence_penalty,
             frequency_penalty=frequency_penalty,
             n=n,
+            prediction=utils.get_pydantic_model(
+                prediction, Optional[models.Prediction]
+            ),
             agent_id=agent_id,
         )
 
@@ -138,13 +145,16 @@ class Agents(BaseSDK):
 
         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, 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"], "*"):
+        if utils.match_response(http_res, "4XX", "*"):
+            http_res_text = utils.stream_to_text(http_res)
+            raise models.SDKError(
+                "API error occurred", http_res.status_code, http_res_text, http_res
+            )
+        if utils.match_response(http_res, "5XX", "*"):
             http_res_text = utils.stream_to_text(http_res)
             raise models.SDKError(
                 "API error occurred", http_res.status_code, http_res_text, http_res
@@ -191,11 +201,14 @@ class Agents(BaseSDK):
         presence_penalty: Optional[float] = None,
         frequency_penalty: Optional[float] = None,
         n: OptionalNullable[int] = UNSET,
+        prediction: Optional[
+            Union[models.Prediction, models.PredictionTypedDict]
+        ] = None,
         retries: OptionalNullable[utils.RetryConfig] = UNSET,
         server_url: Optional[str] = None,
         timeout_ms: Optional[int] = None,
         http_headers: Optional[Mapping[str, str]] = None,
-    ) -> Optional[models.ChatCompletionResponse]:
+    ) -> models.ChatCompletionResponse:
         r"""Agents Completion
 
         :param messages: The prompt(s) to generate completions for, encoded as a list of dict with role and content.
@@ -210,6 +223,7 @@ class Agents(BaseSDK):
         :param presence_penalty: presence_penalty determines how much the model penalizes the repetition of words or phrases. A higher presence penalty encourages the model to use a wider variety of words and phrases, making the output more diverse and creative.
         :param frequency_penalty: frequency_penalty penalizes the repetition of words based on their frequency in the generated text. A higher frequency penalty discourages the model from repeating words that have already appeared frequently in the output, promoting diversity and reducing repetition.
         :param n: Number of completions to return for each request, input tokens are only billed once.
+        :param prediction:
         :param retries: Override the default retry configuration for this method
         :param server_url: Override the default server URL for this method
         :param timeout_ms: Override the default request timeout configuration for this method in milliseconds
@@ -241,6 +255,9 @@ class Agents(BaseSDK):
             presence_penalty=presence_penalty,
             frequency_penalty=frequency_penalty,
             n=n,
+            prediction=utils.get_pydantic_model(
+                prediction, Optional[models.Prediction]
+            ),
             agent_id=agent_id,
         )
 
@@ -286,13 +303,16 @@ class Agents(BaseSDK):
 
         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, 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"], "*"):
+        if utils.match_response(http_res, "4XX", "*"):
+            http_res_text = await utils.stream_to_text_async(http_res)
+            raise models.SDKError(
+                "API error occurred", http_res.status_code, http_res_text, http_res
+            )
+        if utils.match_response(http_res, "5XX", "*"):
             http_res_text = await utils.stream_to_text_async(http_res)
             raise models.SDKError(
                 "API error occurred", http_res.status_code, http_res_text, http_res
@@ -339,11 +359,14 @@ class Agents(BaseSDK):
         presence_penalty: Optional[float] = None,
         frequency_penalty: Optional[float] = None,
         n: OptionalNullable[int] = UNSET,
+        prediction: Optional[
+            Union[models.Prediction, models.PredictionTypedDict]
+        ] = None,
         retries: OptionalNullable[utils.RetryConfig] = UNSET,
         server_url: Optional[str] = None,
         timeout_ms: Optional[int] = None,
         http_headers: Optional[Mapping[str, str]] = None,
-    ) -> Optional[eventstreaming.EventStream[models.CompletionEvent]]:
+    ) -> eventstreaming.EventStream[models.CompletionEvent]:
         r"""Stream Agents completion
 
         Mistral AI provides the ability to stream responses back to a client in order to allow partial results for certain requests. Tokens will be sent as data-only server-sent 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.
@@ -360,6 +383,7 @@ class Agents(BaseSDK):
         :param presence_penalty: presence_penalty determines how much the model penalizes the repetition of words or phrases. A higher presence penalty encourages the model to use a wider variety of words and phrases, making the output more diverse and creative.
         :param frequency_penalty: frequency_penalty penalizes the repetition of words based on their frequency in the generated text. A higher frequency penalty discourages the model from repeating words that have already appeared frequently in the output, promoting diversity and reducing repetition.
         :param n: Number of completions to return for each request, input tokens are only billed once.
+        :param prediction:
         :param retries: Override the default retry configuration for this method
         :param server_url: Override the default server URL for this method
         :param timeout_ms: Override the default request timeout configuration for this method in milliseconds
@@ -391,6 +415,9 @@ class Agents(BaseSDK):
             presence_penalty=presence_penalty,
             frequency_penalty=frequency_penalty,
             n=n,
+            prediction=utils.get_pydantic_model(
+                prediction, Optional[models.Prediction]
+            ),
             agent_id=agent_id,
         )
 
@@ -446,7 +473,12 @@ class Agents(BaseSDK):
             http_res_text = utils.stream_to_text(http_res)
             data = utils.unmarshal_json(http_res_text, models.HTTPValidationErrorData)
             raise models.HTTPValidationError(data=data)
-        if utils.match_response(http_res, ["4XX", "5XX"], "*"):
+        if utils.match_response(http_res, "4XX", "*"):
+            http_res_text = utils.stream_to_text(http_res)
+            raise models.SDKError(
+                "API error occurred", http_res.status_code, http_res_text, http_res
+            )
+        if utils.match_response(http_res, "5XX", "*"):
             http_res_text = utils.stream_to_text(http_res)
             raise models.SDKError(
                 "API error occurred", http_res.status_code, http_res_text, http_res
@@ -493,11 +525,14 @@ class Agents(BaseSDK):
         presence_penalty: Optional[float] = None,
         frequency_penalty: Optional[float] = None,
         n: OptionalNullable[int] = UNSET,
+        prediction: Optional[
+            Union[models.Prediction, models.PredictionTypedDict]
+        ] = None,
         retries: OptionalNullable[utils.RetryConfig] = UNSET,
         server_url: Optional[str] = None,
         timeout_ms: Optional[int] = None,
         http_headers: Optional[Mapping[str, str]] = None,
-    ) -> Optional[eventstreaming.EventStreamAsync[models.CompletionEvent]]:
+    ) -> eventstreaming.EventStreamAsync[models.CompletionEvent]:
         r"""Stream Agents completion
 
         Mistral AI provides the ability to stream responses back to a client in order to allow partial results for certain requests. Tokens will be sent as data-only server-sent 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.
@@ -514,6 +549,7 @@ class Agents(BaseSDK):
         :param presence_penalty: presence_penalty determines how much the model penalizes the repetition of words or phrases. A higher presence penalty encourages the model to use a wider variety of words and phrases, making the output more diverse and creative.
         :param frequency_penalty: frequency_penalty penalizes the repetition of words based on their frequency in the generated text. A higher frequency penalty discourages the model from repeating words that have already appeared frequently in the output, promoting diversity and reducing repetition.
         :param n: Number of completions to return for each request, input tokens are only billed once.
+        :param prediction:
         :param retries: Override the default retry configuration for this method
         :param server_url: Override the default server URL for this method
         :param timeout_ms: Override the default request timeout configuration for this method in milliseconds
@@ -545,6 +581,9 @@ class Agents(BaseSDK):
             presence_penalty=presence_penalty,
             frequency_penalty=frequency_penalty,
             n=n,
+            prediction=utils.get_pydantic_model(
+                prediction, Optional[models.Prediction]
+            ),
             agent_id=agent_id,
         )
 
@@ -600,7 +639,12 @@ class Agents(BaseSDK):
             http_res_text = await utils.stream_to_text_async(http_res)
             data = utils.unmarshal_json(http_res_text, models.HTTPValidationErrorData)
             raise models.HTTPValidationError(data=data)
-        if utils.match_response(http_res, ["4XX", "5XX"], "*"):
+        if utils.match_response(http_res, "4XX", "*"):
+            http_res_text = await utils.stream_to_text_async(http_res)
+            raise models.SDKError(
+                "API error occurred", http_res.status_code, http_res_text, http_res
+            )
+        if utils.match_response(http_res, "5XX", "*"):
             http_res_text = await utils.stream_to_text_async(http_res)
             raise models.SDKError(
                 "API error occurred", http_res.status_code, http_res_text, http_res

mistralai/chat.py

@@ -7,10 +7,92 @@ from mistralai.types import Nullable, OptionalNullable, UNSET
 from mistralai.utils import eventstreaming, get_security_from_env
 from typing import Any, List, Mapping, Optional, Union
 
+# region imports
+from typing import Type
+from mistralai.extra import (
+    convert_to_parsed_chat_completion_response,
+    response_format_from_pydantic_model,
+    CustomPydanticModel,
+    ParsedChatCompletionResponse,
+)
+# endregion imports
+
 
 class Chat(BaseSDK):
     r"""Chat Completion API."""
 
+    # region sdk-class-body
+    # Custom .parse methods for the Structure Outputs Feature.
+
+    def parse(
+        self, response_format: Type[CustomPydanticModel], **kwargs: Any
+    ) -> ParsedChatCompletionResponse[CustomPydanticModel]:
+        """
+        Parse the response using the provided response format.
+        :param Type[CustomPydanticModel] response_format: The Pydantic model to parse the response into
+        :param Any **kwargs Additional keyword arguments to pass to the .complete method
+        :return: The parsed response
+        """
+        # Convert the input Pydantic Model to a strict JSON ready to be passed to chat.complete
+        json_response_format = response_format_from_pydantic_model(response_format)
+        # Run the inference
+        response = self.complete(**kwargs, response_format=json_response_format)
+        # Parse response back to the input pydantic model
+        parsed_response = convert_to_parsed_chat_completion_response(
+            response, response_format
+        )
+        return parsed_response
+
+    async def parse_async(
+        self, response_format: Type[CustomPydanticModel], **kwargs
+    ) -> ParsedChatCompletionResponse[CustomPydanticModel]:
+        """
+        Asynchronously parse the response using the provided response format.
+        :param Type[CustomPydanticModel] response_format: The Pydantic model to parse the response into
+        :param Any **kwargs Additional keyword arguments to pass to the .complete method
+        :return: The parsed response
+        """
+        json_response_format = response_format_from_pydantic_model(response_format)
+        response = await self.complete_async(  # pylint: disable=E1125
+            **kwargs, response_format=json_response_format
+        )
+        parsed_response = convert_to_parsed_chat_completion_response(
+            response, response_format
+        )
+        return parsed_response
+
+    def parse_stream(
+        self, response_format: Type[CustomPydanticModel], **kwargs
+    ) -> eventstreaming.EventStream[models.CompletionEvent]:
+        """
+        Parse the response using the provided response format.
+        For now the response will be in JSON format not in the input Pydantic model.
+        :param Type[CustomPydanticModel] response_format: The Pydantic model to parse the response into
+        :param Any **kwargs Additional keyword arguments to pass to the .stream method
+        :return: The JSON parsed response
+        """
+        json_response_format = response_format_from_pydantic_model(response_format)
+        response = self.stream(**kwargs, response_format=json_response_format)
+        return response
+
+    async def parse_stream_async(
+        self, response_format: Type[CustomPydanticModel], **kwargs
+    ) -> eventstreaming.EventStreamAsync[models.CompletionEvent]:
+        """
+        Asynchronously parse the response using the provided response format.
+        For now the response will be in JSON format not in the input Pydantic model.
+        :param Type[CustomPydanticModel] response_format: The Pydantic model to parse the response into
+        :param Any **kwargs Additional keyword arguments to pass to the .stream method
+        :return: The JSON parsed response
+        """
+        json_response_format = response_format_from_pydantic_model(response_format)
+        response = await self.stream_async(  # pylint: disable=E1125
+            **kwargs, response_format=json_response_format
+        )
+        return response
+
+    # endregion sdk-class-body
+
     def complete(
         self,
         *,
@@ -37,12 +119,15 @@ class Chat(BaseSDK):
         presence_penalty: Optional[float] = None,
         frequency_penalty: Optional[float] = None,
         n: OptionalNullable[int] = UNSET,
+        prediction: Optional[
+            Union[models.Prediction, models.PredictionTypedDict]
+        ] = None,
         safe_prompt: Optional[bool] = None,
         retries: OptionalNullable[utils.RetryConfig] = UNSET,
         server_url: Optional[str] = None,
         timeout_ms: Optional[int] = None,
         http_headers: Optional[Mapping[str, str]] = None,
-    ) -> Optional[models.ChatCompletionResponse]:
+    ) -> models.ChatCompletionResponse:
         r"""Chat Completion
 
         :param model: ID of the model to use. You can use the [List Available Models](/api/#tag/models/operation/list_models_v1_models_get) API to see all of your available models, or see our [Model overview](/models) for model descriptions.
@@ -59,6 +144,7 @@ class Chat(BaseSDK):
         :param presence_penalty: presence_penalty determines how much the model penalizes the repetition of words or phrases. A higher presence penalty encourages the model to use a wider variety of words and phrases, making the output more diverse and creative.
         :param frequency_penalty: frequency_penalty penalizes the repetition of words based on their frequency in the generated text. A higher frequency penalty discourages the model from repeating words that have already appeared frequently in the output, promoting diversity and reducing repetition.
         :param n: Number of completions to return for each request, input tokens are only billed once.
+        :param prediction:
         :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
@@ -92,6 +178,9 @@ class Chat(BaseSDK):
             presence_penalty=presence_penalty,
             frequency_penalty=frequency_penalty,
             n=n,
+            prediction=utils.get_pydantic_model(
+                prediction, Optional[models.Prediction]
+            ),
             safe_prompt=safe_prompt,
         )
 
@@ -137,13 +226,16 @@ class Chat(BaseSDK):
 
         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, 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"], "*"):
+        if utils.match_response(http_res, "4XX", "*"):
+            http_res_text = utils.stream_to_text(http_res)
+            raise models.SDKError(
+                "API error occurred", http_res.status_code, http_res_text, http_res
+            )
+        if utils.match_response(http_res, "5XX", "*"):
             http_res_text = utils.stream_to_text(http_res)
             raise models.SDKError(
                 "API error occurred", http_res.status_code, http_res_text, http_res
@@ -184,12 +276,15 @@ class Chat(BaseSDK):
         presence_penalty: Optional[float] = None,
         frequency_penalty: Optional[float] = None,
         n: OptionalNullable[int] = UNSET,
+        prediction: Optional[
+            Union[models.Prediction, models.PredictionTypedDict]
+        ] = None,
         safe_prompt: Optional[bool] = None,
         retries: OptionalNullable[utils.RetryConfig] = UNSET,
         server_url: Optional[str] = None,
         timeout_ms: Optional[int] = None,
         http_headers: Optional[Mapping[str, str]] = None,
-    ) -> Optional[models.ChatCompletionResponse]:
+    ) -> models.ChatCompletionResponse:
         r"""Chat Completion
 
         :param model: ID of the model to use. You can use the [List Available Models](/api/#tag/models/operation/list_models_v1_models_get) API to see all of your available models, or see our [Model overview](/models) for model descriptions.
@@ -206,6 +301,7 @@ class Chat(BaseSDK):
         :param presence_penalty: presence_penalty determines how much the model penalizes the repetition of words or phrases. A higher presence penalty encourages the model to use a wider variety of words and phrases, making the output more diverse and creative.
         :param frequency_penalty: frequency_penalty penalizes the repetition of words based on their frequency in the generated text. A higher frequency penalty discourages the model from repeating words that have already appeared frequently in the output, promoting diversity and reducing repetition.
         :param n: Number of completions to return for each request, input tokens are only billed once.
+        :param prediction:
         :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
@@ -239,6 +335,9 @@ class Chat(BaseSDK):
             presence_penalty=presence_penalty,
             frequency_penalty=frequency_penalty,
             n=n,
+            prediction=utils.get_pydantic_model(
+                prediction, Optional[models.Prediction]
+            ),
             safe_prompt=safe_prompt,
         )
 
@@ -284,13 +383,16 @@ class Chat(BaseSDK):
 
         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, 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"], "*"):
+        if utils.match_response(http_res, "4XX", "*"):
+            http_res_text = await utils.stream_to_text_async(http_res)
+            raise models.SDKError(
+                "API error occurred", http_res.status_code, http_res_text, http_res
+            )
+        if utils.match_response(http_res, "5XX", "*"):
             http_res_text = await utils.stream_to_text_async(http_res)
             raise models.SDKError(
                 "API error occurred", http_res.status_code, http_res_text, http_res
@@ -339,12 +441,15 @@ class Chat(BaseSDK):
         presence_penalty: Optional[float] = None,
         frequency_penalty: Optional[float] = None,
         n: OptionalNullable[int] = UNSET,
+        prediction: Optional[
+            Union[models.Prediction, models.PredictionTypedDict]
+        ] = None,
         safe_prompt: Optional[bool] = None,
         retries: OptionalNullable[utils.RetryConfig] = UNSET,
         server_url: Optional[str] = None,
         timeout_ms: Optional[int] = None,
         http_headers: Optional[Mapping[str, str]] = None,
-    ) -> Optional[eventstreaming.EventStream[models.CompletionEvent]]:
+    ) -> eventstreaming.EventStream[models.CompletionEvent]:
         r"""Stream chat completion
 
         Mistral AI provides the ability to stream responses back to a client in order to allow partial results for certain requests. Tokens will be sent as data-only server-sent 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.
@@ -363,6 +468,7 @@ class Chat(BaseSDK):
         :param presence_penalty: presence_penalty determines how much the model penalizes the repetition of words or phrases. A higher presence penalty encourages the model to use a wider variety of words and phrases, making the output more diverse and creative.
         :param frequency_penalty: frequency_penalty penalizes the repetition of words based on their frequency in the generated text. A higher frequency penalty discourages the model from repeating words that have already appeared frequently in the output, promoting diversity and reducing repetition.
         :param n: Number of completions to return for each request, input tokens are only billed once.
+        :param prediction:
         :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
@@ -398,6 +504,9 @@ class Chat(BaseSDK):
             presence_penalty=presence_penalty,
             frequency_penalty=frequency_penalty,
             n=n,
+            prediction=utils.get_pydantic_model(
+                prediction, Optional[models.Prediction]
+            ),
             safe_prompt=safe_prompt,
         )
 
@@ -453,7 +562,12 @@ class Chat(BaseSDK):
             http_res_text = utils.stream_to_text(http_res)
             data = utils.unmarshal_json(http_res_text, models.HTTPValidationErrorData)
             raise models.HTTPValidationError(data=data)
-        if utils.match_response(http_res, ["4XX", "5XX"], "*"):
+        if utils.match_response(http_res, "4XX", "*"):
+            http_res_text = utils.stream_to_text(http_res)
+            raise models.SDKError(
+                "API error occurred", http_res.status_code, http_res_text, http_res
+            )
+        if utils.match_response(http_res, "5XX", "*"):
             http_res_text = utils.stream_to_text(http_res)
             raise models.SDKError(
                 "API error occurred", http_res.status_code, http_res_text, http_res
@@ -502,12 +616,15 @@ class Chat(BaseSDK):
         presence_penalty: Optional[float] = None,
         frequency_penalty: Optional[float] = None,
         n: OptionalNullable[int] = UNSET,
+        prediction: Optional[
+            Union[models.Prediction, models.PredictionTypedDict]
+        ] = None,
         safe_prompt: Optional[bool] = None,
         retries: OptionalNullable[utils.RetryConfig] = UNSET,
         server_url: Optional[str] = None,
         timeout_ms: Optional[int] = None,
         http_headers: Optional[Mapping[str, str]] = None,
-    ) -> Optional[eventstreaming.EventStreamAsync[models.CompletionEvent]]:
+    ) -> eventstreaming.EventStreamAsync[models.CompletionEvent]:
         r"""Stream chat completion
 
         Mistral AI provides the ability to stream responses back to a client in order to allow partial results for certain requests. Tokens will be sent as data-only server-sent 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.
@@ -526,6 +643,7 @@ class Chat(BaseSDK):
         :param presence_penalty: presence_penalty determines how much the model penalizes the repetition of words or phrases. A higher presence penalty encourages the model to use a wider variety of words and phrases, making the output more diverse and creative.
         :param frequency_penalty: frequency_penalty penalizes the repetition of words based on their frequency in the generated text. A higher frequency penalty discourages the model from repeating words that have already appeared frequently in the output, promoting diversity and reducing repetition.
         :param n: Number of completions to return for each request, input tokens are only billed once.
+        :param prediction:
         :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
@@ -561,6 +679,9 @@ class Chat(BaseSDK):
             presence_penalty=presence_penalty,
             frequency_penalty=frequency_penalty,
             n=n,
+            prediction=utils.get_pydantic_model(
+                prediction, Optional[models.Prediction]
+            ),
             safe_prompt=safe_prompt,
         )
 
@@ -616,7 +737,12 @@ class Chat(BaseSDK):
             http_res_text = await utils.stream_to_text_async(http_res)
             data = utils.unmarshal_json(http_res_text, models.HTTPValidationErrorData)
             raise models.HTTPValidationError(data=data)
-        if utils.match_response(http_res, ["4XX", "5XX"], "*"):
+        if utils.match_response(http_res, "4XX", "*"):
+            http_res_text = await utils.stream_to_text_async(http_res)
+            raise models.SDKError(
+                "API error occurred", http_res.status_code, http_res_text, http_res
+            )
+        if utils.match_response(http_res, "5XX", "*"):
             http_res_text = await utils.stream_to_text_async(http_res)
             raise models.SDKError(
                 "API error occurred", http_res.status_code, http_res_text, http_res

mistralai/classifiers.py

@@ -23,7 +23,7 @@ class Classifiers(BaseSDK):
         server_url: Optional[str] = None,
         timeout_ms: Optional[int] = None,
         http_headers: Optional[Mapping[str, str]] = None,
-    ) -> Optional[models.ClassificationResponse]:
+    ) -> models.ClassificationResponse:
         r"""Moderations
 
         :param inputs: Text to classify.
@@ -88,13 +88,16 @@ class Classifiers(BaseSDK):
 
         data: Any = None
         if utils.match_response(http_res, "200", "application/json"):
-            return utils.unmarshal_json(
-                http_res.text, Optional[models.ClassificationResponse]
-            )
+            return utils.unmarshal_json(http_res.text, models.ClassificationResponse)
         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"], "*"):
+        if utils.match_response(http_res, "4XX", "*"):
+            http_res_text = utils.stream_to_text(http_res)
+            raise models.SDKError(
+                "API error occurred", http_res.status_code, http_res_text, http_res
+            )
+        if utils.match_response(http_res, "5XX", "*"):
             http_res_text = utils.stream_to_text(http_res)
             raise models.SDKError(
                 "API error occurred", http_res.status_code, http_res_text, http_res
@@ -121,7 +124,7 @@ class Classifiers(BaseSDK):
         server_url: Optional[str] = None,
         timeout_ms: Optional[int] = None,
         http_headers: Optional[Mapping[str, str]] = None,
-    ) -> Optional[models.ClassificationResponse]:
+    ) -> models.ClassificationResponse:
         r"""Moderations
 
         :param inputs: Text to classify.
@@ -186,13 +189,16 @@ class Classifiers(BaseSDK):
 
         data: Any = None
         if utils.match_response(http_res, "200", "application/json"):
-            return utils.unmarshal_json(
-                http_res.text, Optional[models.ClassificationResponse]
-            )
+            return utils.unmarshal_json(http_res.text, models.ClassificationResponse)
         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"], "*"):
+        if utils.match_response(http_res, "4XX", "*"):
+            http_res_text = await utils.stream_to_text_async(http_res)
+            raise models.SDKError(
+                "API error occurred", http_res.status_code, http_res_text, http_res
+            )
+        if utils.match_response(http_res, "5XX", "*"):
             http_res_text = await utils.stream_to_text_async(http_res)
             raise models.SDKError(
                 "API error occurred", http_res.status_code, http_res_text, http_res
@@ -219,7 +225,7 @@ class Classifiers(BaseSDK):
         server_url: Optional[str] = None,
         timeout_ms: Optional[int] = None,
         http_headers: Optional[Mapping[str, str]] = None,
-    ) -> Optional[models.ClassificationResponse]:
+    ) -> models.ClassificationResponse:
         r"""Moderations Chat
 
         :param inputs: Chat to classify
@@ -286,13 +292,16 @@ class Classifiers(BaseSDK):
 
         data: Any = None
         if utils.match_response(http_res, "200", "application/json"):
-            return utils.unmarshal_json(
-                http_res.text, Optional[models.ClassificationResponse]
-            )
+            return utils.unmarshal_json(http_res.text, models.ClassificationResponse)
         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"], "*"):
+        if utils.match_response(http_res, "4XX", "*"):
+            http_res_text = utils.stream_to_text(http_res)
+            raise models.SDKError(
+                "API error occurred", http_res.status_code, http_res_text, http_res
+            )
+        if utils.match_response(http_res, "5XX", "*"):
             http_res_text = utils.stream_to_text(http_res)
             raise models.SDKError(
                 "API error occurred", http_res.status_code, http_res_text, http_res
@@ -319,7 +328,7 @@ class Classifiers(BaseSDK):
         server_url: Optional[str] = None,
         timeout_ms: Optional[int] = None,
         http_headers: Optional[Mapping[str, str]] = None,
-    ) -> Optional[models.ClassificationResponse]:
+    ) -> models.ClassificationResponse:
         r"""Moderations Chat
 
         :param inputs: Chat to classify
@@ -386,13 +395,16 @@ class Classifiers(BaseSDK):
 
         data: Any = None
         if utils.match_response(http_res, "200", "application/json"):
-            return utils.unmarshal_json(
-                http_res.text, Optional[models.ClassificationResponse]
-            )
+            return utils.unmarshal_json(http_res.text, models.ClassificationResponse)
         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"], "*"):
+        if utils.match_response(http_res, "4XX", "*"):
+            http_res_text = await utils.stream_to_text_async(http_res)
+            raise models.SDKError(
+                "API error occurred", http_res.status_code, http_res_text, http_res
+            )
+        if utils.match_response(http_res, "5XX", "*"):
             http_res_text = await utils.stream_to_text_async(http_res)
             raise models.SDKError(
                 "API error occurred", http_res.status_code, http_res_text, http_res