mistralai 0.5.5a50__py3-none-any.whl → 1.0.0__py3-none-any.whl

mistralai_gcp/chat.py

@@ -1,10 +1,9 @@
 """Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT."""
 
 from .basesdk import BaseSDK
-from mistralai_gcp import models
+from mistralai_gcp import models, utils
 from mistralai_gcp._hooks import HookContext
-from mistralai_gcp.types import OptionalNullable, UNSET
-import mistralai_gcp.utils as utils
+from mistralai_gcp.types import Nullable, OptionalNullable, UNSET
 from mistralai_gcp.utils import eventstreaming
 from typing import Any, AsyncGenerator, Generator, List, Optional, Union
 
@@ -14,8 +13,8 @@ class Chat(BaseSDK):
     
     def stream(
         self, *,
+        model: Nullable[str],
         messages: Union[List[models.Messages], List[models.MessagesTypedDict]],
-        model: OptionalNullable[str] = UNSET,
         temperature: Optional[float] = 0.7,
         top_p: Optional[float] = 1,
         max_tokens: OptionalNullable[int] = UNSET,
@@ -34,8 +33,8 @@ class Chat(BaseSDK):
 
         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.
 
-        :param messages: The prompt(s) to generate completions for, encoded as a list of dict with role and content.
         :param model: ID of the model to use. You can use the [List Available Models](/api#operation/listModels) API to see all of your available models, or see our [Model overview](/models) for model descriptions.
+        :param messages: The prompt(s) to generate completions for, encoded as a list of dict with role and content.
         :param temperature: What sampling temperature to use, between 0.0 and 1.0. 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.
         :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.
@@ -127,8 +126,8 @@ class Chat(BaseSDK):
     
     async def stream_async(
         self, *,
+        model: Nullable[str],
         messages: Union[List[models.Messages], List[models.MessagesTypedDict]],
-        model: OptionalNullable[str] = UNSET,
         temperature: Optional[float] = 0.7,
         top_p: Optional[float] = 1,
         max_tokens: OptionalNullable[int] = UNSET,
@@ -147,8 +146,8 @@ class Chat(BaseSDK):
 
         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.
 
-        :param messages: The prompt(s) to generate completions for, encoded as a list of dict with role and content.
         :param model: ID of the model to use. You can use the [List Available Models](/api#operation/listModels) API to see all of your available models, or see our [Model overview](/models) for model descriptions.
+        :param messages: The prompt(s) to generate completions for, encoded as a list of dict with role and content.
         :param temperature: What sampling temperature to use, between 0.0 and 1.0. 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.
         :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.
@@ -238,10 +237,10 @@ class Chat(BaseSDK):
 
     
     
-    def create(
+    def complete(
         self, *,
+        model: Nullable[str],
         messages: Union[List[models.ChatCompletionRequestMessages], List[models.ChatCompletionRequestMessagesTypedDict]],
-        model: OptionalNullable[str] = UNSET,
         temperature: Optional[float] = 0.7,
         top_p: Optional[float] = 1,
         max_tokens: OptionalNullable[int] = UNSET,
@@ -258,10 +257,8 @@ class Chat(BaseSDK):
     ) -> Optional[models.ChatCompletionResponse]:
         r"""Chat Completion
 
-        Chat Completion
-
-        :param messages: The prompt(s) to generate completions for, encoded as a list of dict with role and content.
         :param model: ID of the model to use. You can use the [List Available Models](/api#operation/listModels) API to see all of your available models, or see our [Model overview](/models) for model descriptions.
+        :param messages: The prompt(s) to generate completions for, encoded as a list of dict with role and content.
         :param temperature: What sampling temperature to use, between 0.0 and 1.0. 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.
         :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.
@@ -350,10 +347,10 @@ class Chat(BaseSDK):
 
     
     
-    async def create_async(
+    async def complete_async(
         self, *,
+        model: Nullable[str],
         messages: Union[List[models.ChatCompletionRequestMessages], List[models.ChatCompletionRequestMessagesTypedDict]],
-        model: OptionalNullable[str] = UNSET,
         temperature: Optional[float] = 0.7,
         top_p: Optional[float] = 1,
         max_tokens: OptionalNullable[int] = UNSET,
@@ -370,10 +367,8 @@ class Chat(BaseSDK):
     ) -> Optional[models.ChatCompletionResponse]:
         r"""Chat Completion
 
-        Chat Completion
-
-        :param messages: The prompt(s) to generate completions for, encoded as a list of dict with role and content.
         :param model: ID of the model to use. You can use the [List Available Models](/api#operation/listModels) API to see all of your available models, or see our [Model overview](/models) for model descriptions.
+        :param messages: The prompt(s) to generate completions for, encoded as a list of dict with role and content.
         :param temperature: What sampling temperature to use, between 0.0 and 1.0. 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.
         :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.

mistralai_gcp/fim.py

@@ -1,10 +1,9 @@
 """Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT."""
 
 from .basesdk import BaseSDK
-from mistralai_gcp import models
+from mistralai_gcp import models, utils
 from mistralai_gcp._hooks import HookContext
-from mistralai_gcp.types import OptionalNullable, UNSET
-import mistralai_gcp.utils as utils
+from mistralai_gcp.types import Nullable, OptionalNullable, UNSET
 from mistralai_gcp.utils import eventstreaming
 from typing import Any, AsyncGenerator, Generator, Optional, Union
 
@@ -14,8 +13,8 @@ class Fim(BaseSDK):
     
     def stream(
         self, *,
+        model: Nullable[str],
         prompt: str,
-        model: OptionalNullable[str] = UNSET,
         temperature: Optional[float] = 0.7,
         top_p: Optional[float] = 1,
         max_tokens: OptionalNullable[int] = UNSET,
@@ -32,8 +31,8 @@ class Fim(BaseSDK):
 
         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.
 
-        :param prompt: The text/code to complete.
         :param model: ID of the model to use. Only compatible for now with:   - `codestral-2405`   - `codestral-latest`
+        :param prompt: The text/code to complete.
         :param temperature: What sampling temperature to use, between 0.0 and 1.0. 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.
         :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.
@@ -121,8 +120,8 @@ class Fim(BaseSDK):
     
     async def stream_async(
         self, *,
+        model: Nullable[str],
         prompt: str,
-        model: OptionalNullable[str] = UNSET,
         temperature: Optional[float] = 0.7,
         top_p: Optional[float] = 1,
         max_tokens: OptionalNullable[int] = UNSET,
@@ -139,8 +138,8 @@ class Fim(BaseSDK):
 
         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.
 
-        :param prompt: The text/code to complete.
         :param model: ID of the model to use. Only compatible for now with:   - `codestral-2405`   - `codestral-latest`
+        :param prompt: The text/code to complete.
         :param temperature: What sampling temperature to use, between 0.0 and 1.0. 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.
         :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.
@@ -226,10 +225,10 @@ class Fim(BaseSDK):
 
     
     
-    def create(
+    def complete(
         self, *,
+        model: Nullable[str],
         prompt: str,
-        model: OptionalNullable[str] = UNSET,
         temperature: Optional[float] = 0.7,
         top_p: Optional[float] = 1,
         max_tokens: OptionalNullable[int] = UNSET,
@@ -246,8 +245,8 @@ class Fim(BaseSDK):
 
         FIM completion.
 
-        :param prompt: The text/code to complete.
         :param model: ID of the model to use. Only compatible for now with:   - `codestral-2405`   - `codestral-latest`
+        :param prompt: The text/code to complete.
         :param temperature: What sampling temperature to use, between 0.0 and 1.0. 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.
         :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.
@@ -332,10 +331,10 @@ class Fim(BaseSDK):
 
     
     
-    async def create_async(
+    async def complete_async(
         self, *,
+        model: Nullable[str],
         prompt: str,
-        model: OptionalNullable[str] = UNSET,
         temperature: Optional[float] = 0.7,
         top_p: Optional[float] = 1,
         max_tokens: OptionalNullable[int] = UNSET,
@@ -352,8 +351,8 @@ class Fim(BaseSDK):
 
         FIM completion.
 
-        :param prompt: The text/code to complete.
         :param model: ID of the model to use. Only compatible for now with:   - `codestral-2405`   - `codestral-latest`
+        :param prompt: The text/code to complete.
         :param temperature: What sampling temperature to use, between 0.0 and 1.0. 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.
         :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.

mistralai_gcp/models/__init__.py

@@ -1,7 +1,7 @@
 """Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT."""
 
 from .assistantmessage import AssistantMessage, AssistantMessageRole, AssistantMessageTypedDict
-from .chatcompletionchoice import ChatCompletionChoice, ChatCompletionChoiceFinishReason, ChatCompletionChoiceTypedDict, Message, MessageTypedDict
+from .chatcompletionchoice import ChatCompletionChoice, ChatCompletionChoiceFinishReason, ChatCompletionChoiceTypedDict
 from .chatcompletionrequest import ChatCompletionRequest, ChatCompletionRequestMessages, ChatCompletionRequestMessagesTypedDict, ChatCompletionRequestStop, ChatCompletionRequestStopTypedDict, ChatCompletionRequestToolChoice, ChatCompletionRequestTypedDict
 from .chatcompletionresponse import ChatCompletionResponse, ChatCompletionResponseTypedDict
 from .chatcompletionstreamrequest import ChatCompletionStreamRequest, ChatCompletionStreamRequestTypedDict, Messages, MessagesTypedDict, Stop, StopTypedDict, ToolChoice
@@ -14,7 +14,7 @@ from .fimcompletionrequest import FIMCompletionRequest, FIMCompletionRequestStop
 from .fimcompletionresponse import FIMCompletionResponse, FIMCompletionResponseTypedDict
 from .fimcompletionstreamrequest import FIMCompletionStreamRequest, FIMCompletionStreamRequestStop, FIMCompletionStreamRequestStopTypedDict, FIMCompletionStreamRequestTypedDict
 from .function import Function, FunctionTypedDict
-from .functioncall import FunctionCall, FunctionCallTypedDict
+from .functioncall import Arguments, ArgumentsTypedDict, FunctionCall, FunctionCallTypedDict
 from .httpvalidationerror import HTTPValidationError, HTTPValidationErrorData
 from .responseformat import ResponseFormat, ResponseFormatTypedDict, ResponseFormats
 from .sdkerror import SDKError
@@ -28,4 +28,4 @@ from .usageinfo import UsageInfo, UsageInfoTypedDict
 from .usermessage import UserMessage, UserMessageContent, UserMessageContentTypedDict, UserMessageRole, UserMessageTypedDict
 from .validationerror import Loc, LocTypedDict, ValidationError, ValidationErrorTypedDict
 
-__all__ = ["AssistantMessage", "AssistantMessageRole", "AssistantMessageTypedDict", "ChatCompletionChoice", "ChatCompletionChoiceFinishReason", "ChatCompletionChoiceTypedDict", "ChatCompletionRequest", "ChatCompletionRequestMessages", "ChatCompletionRequestMessagesTypedDict", "ChatCompletionRequestStop", "ChatCompletionRequestStopTypedDict", "ChatCompletionRequestToolChoice", "ChatCompletionRequestTypedDict", "ChatCompletionResponse", "ChatCompletionResponseTypedDict", "ChatCompletionStreamRequest", "ChatCompletionStreamRequestTypedDict", "CompletionChunk", "CompletionChunkTypedDict", "CompletionEvent", "CompletionEventTypedDict", "CompletionResponseStreamChoice", "CompletionResponseStreamChoiceTypedDict", "Content", "ContentChunk", "ContentChunkTypedDict", "ContentTypedDict", "DeltaMessage", "DeltaMessageTypedDict", "FIMCompletionRequest", "FIMCompletionRequestStop", "FIMCompletionRequestStopTypedDict", "FIMCompletionRequestTypedDict", "FIMCompletionResponse", "FIMCompletionResponseTypedDict", "FIMCompletionStreamRequest", "FIMCompletionStreamRequestStop", "FIMCompletionStreamRequestStopTypedDict", "FIMCompletionStreamRequestTypedDict", "FinishReason", "Function", "FunctionCall", "FunctionCallTypedDict", "FunctionTypedDict", "HTTPValidationError", "HTTPValidationErrorData", "Loc", "LocTypedDict", "Message", "MessageTypedDict", "Messages", "MessagesTypedDict", "ResponseFormat", "ResponseFormatTypedDict", "ResponseFormats", "Role", "SDKError", "Security", "SecurityTypedDict", "Stop", "StopTypedDict", "SystemMessage", "SystemMessageTypedDict", "TextChunk", "TextChunkTypedDict", "Tool", "ToolCall", "ToolCallTypedDict", "ToolChoice", "ToolMessage", "ToolMessageRole", "ToolMessageTypedDict", "ToolTypedDict", "UsageInfo", "UsageInfoTypedDict", "UserMessage", "UserMessageContent", "UserMessageContentTypedDict", "UserMessageRole", "UserMessageTypedDict", "ValidationError", "ValidationErrorTypedDict"]
+__all__ = ["Arguments", "ArgumentsTypedDict", "AssistantMessage", "AssistantMessageRole", "AssistantMessageTypedDict", "ChatCompletionChoice", "ChatCompletionChoiceFinishReason", "ChatCompletionChoiceTypedDict", "ChatCompletionRequest", "ChatCompletionRequestMessages", "ChatCompletionRequestMessagesTypedDict", "ChatCompletionRequestStop", "ChatCompletionRequestStopTypedDict", "ChatCompletionRequestToolChoice", "ChatCompletionRequestTypedDict", "ChatCompletionResponse", "ChatCompletionResponseTypedDict", "ChatCompletionStreamRequest", "ChatCompletionStreamRequestTypedDict", "CompletionChunk", "CompletionChunkTypedDict", "CompletionEvent", "CompletionEventTypedDict", "CompletionResponseStreamChoice", "CompletionResponseStreamChoiceTypedDict", "Content", "ContentChunk", "ContentChunkTypedDict", "ContentTypedDict", "DeltaMessage", "DeltaMessageTypedDict", "FIMCompletionRequest", "FIMCompletionRequestStop", "FIMCompletionRequestStopTypedDict", "FIMCompletionRequestTypedDict", "FIMCompletionResponse", "FIMCompletionResponseTypedDict", "FIMCompletionStreamRequest", "FIMCompletionStreamRequestStop", "FIMCompletionStreamRequestStopTypedDict", "FIMCompletionStreamRequestTypedDict", "FinishReason", "Function", "FunctionCall", "FunctionCallTypedDict", "FunctionTypedDict", "HTTPValidationError", "HTTPValidationErrorData", "Loc", "LocTypedDict", "Messages", "MessagesTypedDict", "ResponseFormat", "ResponseFormatTypedDict", "ResponseFormats", "Role", "SDKError", "Security", "SecurityTypedDict", "Stop", "StopTypedDict", "SystemMessage", "SystemMessageTypedDict", "TextChunk", "TextChunkTypedDict", "Tool", "ToolCall", "ToolCallTypedDict", "ToolChoice", "ToolMessage", "ToolMessageRole", "ToolMessageTypedDict", "ToolTypedDict", "UsageInfo", "UsageInfoTypedDict", "UserMessage", "UserMessageContent", "UserMessageContentTypedDict", "UserMessageRole", "UserMessageTypedDict", "ValidationError", "ValidationErrorTypedDict"]

mistralai_gcp/models/assistantmessage.py

@@ -39,18 +39,13 @@ class AssistantMessage(BaseModel):
             k = f.alias or n
             val = serialized.get(k)
 
+            optional_nullable = k in optional_fields and k in nullable_fields
+            is_set = (self.__pydantic_fields_set__.intersection({n}) or k in null_default_fields) # pylint: disable=no-member
+
             if val is not None and val != UNSET_SENTINEL:
                 m[k] = val
             elif val != UNSET_SENTINEL and (
-                not k in optional_fields
-                or (
-                    k in optional_fields
-                    and k in nullable_fields
-                    and (
-                        self.__pydantic_fields_set__.intersection({n})
-                        or k in null_default_fields
-                    )  # pylint: disable=no-member
-                )
+                not k in optional_fields or (optional_nullable and is_set)
             ):
                 m[k] = val
 

mistralai_gcp/models/chatcompletionchoice.py

@@ -2,14 +2,9 @@
 
 from __future__ import annotations
 from .assistantmessage import AssistantMessage, AssistantMessageTypedDict
-from .systemmessage import SystemMessage, SystemMessageTypedDict
-from .toolmessage import ToolMessage, ToolMessageTypedDict
-from .usermessage import UserMessage, UserMessageTypedDict
 from mistralai_gcp.types import BaseModel
-from mistralai_gcp.utils import get_discriminator
-from pydantic import Discriminator, Tag
-from typing import Literal, Optional, TypedDict, Union
-from typing_extensions import Annotated, NotRequired
+from typing import Literal, Optional, TypedDict
+from typing_extensions import NotRequired
 
 
 ChatCompletionChoiceFinishReason = Literal["stop", "length", "model_length", "error", "tool_calls"]
@@ -17,17 +12,11 @@ ChatCompletionChoiceFinishReason = Literal["stop", "length", "model_length", "er
 class ChatCompletionChoiceTypedDict(TypedDict):
     index: int
     finish_reason: ChatCompletionChoiceFinishReason
-    message: NotRequired[MessageTypedDict]
+    message: NotRequired[AssistantMessageTypedDict]
     
 
 class ChatCompletionChoice(BaseModel):
     index: int
     finish_reason: ChatCompletionChoiceFinishReason
-    message: Optional[Message] = None
+    message: Optional[AssistantMessage] = None
     
-
-MessageTypedDict = Union[SystemMessageTypedDict, UserMessageTypedDict, AssistantMessageTypedDict, ToolMessageTypedDict]
-
-
-Message = Annotated[Union[Annotated[AssistantMessage, Tag("assistant")], Annotated[SystemMessage, Tag("system")], Annotated[ToolMessage, Tag("tool")], Annotated[UserMessage, Tag("user")]], Discriminator(lambda m: get_discriminator(m, "role", "role"))]
-

mistralai_gcp/models/chatcompletionrequest.py

@@ -17,10 +17,10 @@ from typing_extensions import Annotated, NotRequired
 ChatCompletionRequestToolChoice = Literal["auto", "none", "any"]
 
 class ChatCompletionRequestTypedDict(TypedDict):
+    model: Nullable[str]
+    r"""ID of the model to use. You can use the [List Available Models](/api#operation/listModels) API to see all of your available models, or see our [Model overview](/models) for model descriptions."""
     messages: List[ChatCompletionRequestMessagesTypedDict]
     r"""The prompt(s) to generate completions for, encoded as a list of dict with role and content."""
-    model: NotRequired[Nullable[str]]
-    r"""ID of the model to use. You can use the [List Available Models](/api#operation/listModels) API to see all of your available models, or see our [Model overview](/models) for model descriptions."""
     temperature: NotRequired[float]
     r"""What sampling temperature to use, between 0.0 and 1.0. 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: NotRequired[float]
@@ -41,10 +41,10 @@ class ChatCompletionRequestTypedDict(TypedDict):
     
 
 class ChatCompletionRequest(BaseModel):
+    model: Nullable[str]
+    r"""ID of the model to use. You can use the [List Available Models](/api#operation/listModels) API to see all of your available models, or see our [Model overview](/models) for model descriptions."""
     messages: List[ChatCompletionRequestMessages]
     r"""The prompt(s) to generate completions for, encoded as a list of dict with role and content."""
-    model: OptionalNullable[str] = UNSET
-    r"""ID of the model to use. You can use the [List Available Models](/api#operation/listModels) API to see all of your available models, or see our [Model overview](/models) for model descriptions."""
     temperature: Optional[float] = 0.7
     r"""What sampling temperature to use, between 0.0 and 1.0. 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: Optional[float] = 1
@@ -65,7 +65,7 @@ class ChatCompletionRequest(BaseModel):
     
     @model_serializer(mode="wrap")
     def serialize_model(self, handler):
-        optional_fields = ["model", "temperature", "top_p", "max_tokens", "min_tokens", "stream", "stop", "random_seed", "response_format", "tools", "tool_choice"]
+        optional_fields = ["temperature", "top_p", "max_tokens", "min_tokens", "stream", "stop", "random_seed", "response_format", "tools", "tool_choice"]
         nullable_fields = ["model", "max_tokens", "min_tokens", "random_seed", "tools"]
         null_default_fields = []
 
@@ -77,18 +77,13 @@ class ChatCompletionRequest(BaseModel):
             k = f.alias or n
             val = serialized.get(k)
 
+            optional_nullable = k in optional_fields and k in nullable_fields
+            is_set = (self.__pydantic_fields_set__.intersection({n}) or k in null_default_fields) # pylint: disable=no-member
+
             if val is not None and val != UNSET_SENTINEL:
                 m[k] = val
             elif val != UNSET_SENTINEL and (
-                not k in optional_fields
-                or (
-                    k in optional_fields
-                    and k in nullable_fields
-                    and (
-                        self.__pydantic_fields_set__.intersection({n})
-                        or k in null_default_fields
-                    )  # pylint: disable=no-member
-                )
+                not k in optional_fields or (optional_nullable and is_set)
             ):
                 m[k] = val
 

mistralai_gcp/models/chatcompletionstreamrequest.py

@@ -17,10 +17,10 @@ from typing_extensions import Annotated, NotRequired
 ToolChoice = Literal["auto", "none", "any"]
 
 class ChatCompletionStreamRequestTypedDict(TypedDict):
+    model: Nullable[str]
+    r"""ID of the model to use. You can use the [List Available Models](/api#operation/listModels) API to see all of your available models, or see our [Model overview](/models) for model descriptions."""
     messages: List[MessagesTypedDict]
     r"""The prompt(s) to generate completions for, encoded as a list of dict with role and content."""
-    model: NotRequired[Nullable[str]]
-    r"""ID of the model to use. You can use the [List Available Models](/api#operation/listModels) API to see all of your available models, or see our [Model overview](/models) for model descriptions."""
     temperature: NotRequired[float]
     r"""What sampling temperature to use, between 0.0 and 1.0. 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: NotRequired[float]
@@ -40,10 +40,10 @@ class ChatCompletionStreamRequestTypedDict(TypedDict):
     
 
 class ChatCompletionStreamRequest(BaseModel):
+    model: Nullable[str]
+    r"""ID of the model to use. You can use the [List Available Models](/api#operation/listModels) API to see all of your available models, or see our [Model overview](/models) for model descriptions."""
     messages: List[Messages]
     r"""The prompt(s) to generate completions for, encoded as a list of dict with role and content."""
-    model: OptionalNullable[str] = UNSET
-    r"""ID of the model to use. You can use the [List Available Models](/api#operation/listModels) API to see all of your available models, or see our [Model overview](/models) for model descriptions."""
     temperature: Optional[float] = 0.7
     r"""What sampling temperature to use, between 0.0 and 1.0. 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: Optional[float] = 1
@@ -63,7 +63,7 @@ class ChatCompletionStreamRequest(BaseModel):
     
     @model_serializer(mode="wrap")
     def serialize_model(self, handler):
-        optional_fields = ["model", "temperature", "top_p", "max_tokens", "min_tokens", "stream", "stop", "random_seed", "response_format", "tools", "tool_choice"]
+        optional_fields = ["temperature", "top_p", "max_tokens", "min_tokens", "stream", "stop", "random_seed", "response_format", "tools", "tool_choice"]
         nullable_fields = ["model", "max_tokens", "min_tokens", "random_seed", "tools"]
         null_default_fields = []
 
@@ -75,18 +75,13 @@ class ChatCompletionStreamRequest(BaseModel):
             k = f.alias or n
             val = serialized.get(k)
 
+            optional_nullable = k in optional_fields and k in nullable_fields
+            is_set = (self.__pydantic_fields_set__.intersection({n}) or k in null_default_fields) # pylint: disable=no-member
+
             if val is not None and val != UNSET_SENTINEL:
                 m[k] = val
             elif val != UNSET_SENTINEL and (
-                not k in optional_fields
-                or (
-                    k in optional_fields
-                    and k in nullable_fields
-                    and (
-                        self.__pydantic_fields_set__.intersection({n})
-                        or k in null_default_fields
-                    )  # pylint: disable=no-member
-                )
+                not k in optional_fields or (optional_nullable and is_set)
             ):
                 m[k] = val
 

mistralai_gcp/models/completionresponsestreamchoice.py

@@ -34,18 +34,13 @@ class CompletionResponseStreamChoice(BaseModel):
             k = f.alias or n
             val = serialized.get(k)
 
+            optional_nullable = k in optional_fields and k in nullable_fields
+            is_set = (self.__pydantic_fields_set__.intersection({n}) or k in null_default_fields) # pylint: disable=no-member
+
             if val is not None and val != UNSET_SENTINEL:
                 m[k] = val
             elif val != UNSET_SENTINEL and (
-                not k in optional_fields
-                or (
-                    k in optional_fields
-                    and k in nullable_fields
-                    and (
-                        self.__pydantic_fields_set__.intersection({n})
-                        or k in null_default_fields
-                    )  # pylint: disable=no-member
-                )
+                not k in optional_fields or (optional_nullable and is_set)
             ):
                 m[k] = val
 

mistralai_gcp/models/deltamessage.py

@@ -33,18 +33,13 @@ class DeltaMessage(BaseModel):
             k = f.alias or n
             val = serialized.get(k)
 
+            optional_nullable = k in optional_fields and k in nullable_fields
+            is_set = (self.__pydantic_fields_set__.intersection({n}) or k in null_default_fields) # pylint: disable=no-member
+
             if val is not None and val != UNSET_SENTINEL:
                 m[k] = val
             elif val != UNSET_SENTINEL and (
-                not k in optional_fields
-                or (
-                    k in optional_fields
-                    and k in nullable_fields
-                    and (
-                        self.__pydantic_fields_set__.intersection({n})
-                        or k in null_default_fields
-                    )  # pylint: disable=no-member
-                )
+                not k in optional_fields or (optional_nullable and is_set)
             ):
                 m[k] = val
 

mistralai_gcp/models/fimcompletionrequest.py

@@ -8,13 +8,13 @@ from typing_extensions import NotRequired
 
 
 class FIMCompletionRequestTypedDict(TypedDict):
-    prompt: str
-    r"""The text/code to complete."""
-    model: NotRequired[Nullable[str]]
+    model: Nullable[str]
     r"""ID of the model to use. Only compatible for now with:
     - `codestral-2405`
     - `codestral-latest`
     """
+    prompt: str
+    r"""The text/code to complete."""
     temperature: NotRequired[float]
     r"""What sampling temperature to use, between 0.0 and 1.0. 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: NotRequired[float]
@@ -34,13 +34,13 @@ class FIMCompletionRequestTypedDict(TypedDict):
     
 
 class FIMCompletionRequest(BaseModel):
-    prompt: str
-    r"""The text/code to complete."""
-    model: OptionalNullable[str] = UNSET
+    model: Nullable[str]
     r"""ID of the model to use. Only compatible for now with:
     - `codestral-2405`
     - `codestral-latest`
     """
+    prompt: str
+    r"""The text/code to complete."""
     temperature: Optional[float] = 0.7
     r"""What sampling temperature to use, between 0.0 and 1.0. 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: Optional[float] = 1
@@ -60,7 +60,7 @@ class FIMCompletionRequest(BaseModel):
     
     @model_serializer(mode="wrap")
     def serialize_model(self, handler):
-        optional_fields = ["model", "temperature", "top_p", "max_tokens", "min_tokens", "stream", "stop", "random_seed", "suffix"]
+        optional_fields = ["temperature", "top_p", "max_tokens", "min_tokens", "stream", "stop", "random_seed", "suffix"]
         nullable_fields = ["model", "max_tokens", "min_tokens", "random_seed", "suffix"]
         null_default_fields = []
 
@@ -72,18 +72,13 @@ class FIMCompletionRequest(BaseModel):
             k = f.alias or n
             val = serialized.get(k)
 
+            optional_nullable = k in optional_fields and k in nullable_fields
+            is_set = (self.__pydantic_fields_set__.intersection({n}) or k in null_default_fields) # pylint: disable=no-member
+
             if val is not None and val != UNSET_SENTINEL:
                 m[k] = val
             elif val != UNSET_SENTINEL and (
-                not k in optional_fields
-                or (
-                    k in optional_fields
-                    and k in nullable_fields
-                    and (
-                        self.__pydantic_fields_set__.intersection({n})
-                        or k in null_default_fields
-                    )  # pylint: disable=no-member
-                )
+                not k in optional_fields or (optional_nullable and is_set)
             ):
                 m[k] = val
 

mistralai_gcp/models/fimcompletionstreamrequest.py

@@ -8,13 +8,13 @@ from typing_extensions import NotRequired
 
 
 class FIMCompletionStreamRequestTypedDict(TypedDict):
-    prompt: str
-    r"""The text/code to complete."""
-    model: NotRequired[Nullable[str]]
+    model: Nullable[str]
     r"""ID of the model to use. Only compatible for now with:
     - `codestral-2405`
     - `codestral-latest`
     """
+    prompt: str
+    r"""The text/code to complete."""
     temperature: NotRequired[float]
     r"""What sampling temperature to use, between 0.0 and 1.0. 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: NotRequired[float]
@@ -33,13 +33,13 @@ class FIMCompletionStreamRequestTypedDict(TypedDict):
     
 
 class FIMCompletionStreamRequest(BaseModel):
-    prompt: str
-    r"""The text/code to complete."""
-    model: OptionalNullable[str] = UNSET
+    model: Nullable[str]
     r"""ID of the model to use. Only compatible for now with:
     - `codestral-2405`
     - `codestral-latest`
     """
+    prompt: str
+    r"""The text/code to complete."""
     temperature: Optional[float] = 0.7
     r"""What sampling temperature to use, between 0.0 and 1.0. 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: Optional[float] = 1
@@ -58,7 +58,7 @@ class FIMCompletionStreamRequest(BaseModel):
     
     @model_serializer(mode="wrap")
     def serialize_model(self, handler):
-        optional_fields = ["model", "temperature", "top_p", "max_tokens", "min_tokens", "stream", "stop", "random_seed", "suffix"]
+        optional_fields = ["temperature", "top_p", "max_tokens", "min_tokens", "stream", "stop", "random_seed", "suffix"]
         nullable_fields = ["model", "max_tokens", "min_tokens", "random_seed", "suffix"]
         null_default_fields = []
 
@@ -70,18 +70,13 @@ class FIMCompletionStreamRequest(BaseModel):
             k = f.alias or n
             val = serialized.get(k)
 
+            optional_nullable = k in optional_fields and k in nullable_fields
+            is_set = (self.__pydantic_fields_set__.intersection({n}) or k in null_default_fields) # pylint: disable=no-member
+
             if val is not None and val != UNSET_SENTINEL:
                 m[k] = val
             elif val != UNSET_SENTINEL and (
-                not k in optional_fields
-                or (
-                    k in optional_fields
-                    and k in nullable_fields
-                    and (
-                        self.__pydantic_fields_set__.intersection({n})
-                        or k in null_default_fields
-                    )  # pylint: disable=no-member
-                )
+                not k in optional_fields or (optional_nullable and is_set)
             ):
                 m[k] = val
 

mistralai_gcp/models/functioncall.py

@@ -2,15 +2,21 @@
 
 from __future__ import annotations
 from mistralai_gcp.types import BaseModel
-from typing import TypedDict
+from typing import Any, Dict, TypedDict, Union
 
 
 class FunctionCallTypedDict(TypedDict):
     name: str
-    arguments: str
+    arguments: ArgumentsTypedDict
     
 
 class FunctionCall(BaseModel):
     name: str
-    arguments: str
+    arguments: Arguments
     
+
+ArgumentsTypedDict = Union[Dict[str, Any], str]
+
+
+Arguments = Union[Dict[str, Any], str]
+

mistralai_gcp/models/httpvalidationerror.py

@@ -2,8 +2,8 @@
 
 from __future__ import annotations
 from .validationerror import ValidationError
+from mistralai_gcp import utils
 from mistralai_gcp.types import BaseModel
-import mistralai_gcp.utils as utils
 from typing import List, Optional
 
 class HTTPValidationErrorData(BaseModel):