mistralai 1.0.0rc2__py3-none-any.whl → 1.0.2__py3-none-any.whl

mistralai/agents.py

@@ -3,7 +3,7 @@
 from .basesdk import BaseSDK
 from mistralai import models, utils
 from mistralai._hooks import HookContext
-from mistralai.types import Nullable, OptionalNullable, UNSET
+from mistralai.types import OptionalNullable, UNSET
 from mistralai.utils import eventstreaming, get_security_from_env
 from typing import Any, AsyncGenerator, Generator, List, Optional, Union
 
@@ -27,7 +27,7 @@ class Agents(BaseSDK):
         server_url: Optional[str] = None,
         timeout_ms: Optional[int] = None,
     ) -> Optional[models.ChatCompletionResponse]:
-        r"""Chat Completion
+        r"""Agents Completion
 
         :param messages: The prompt(s) to generate completions for, encoded as a list of dict with role and content.
         :param agent_id: The ID of the agent to use for this completion.
@@ -131,7 +131,7 @@ class Agents(BaseSDK):
         server_url: Optional[str] = None,
         timeout_ms: Optional[int] = None,
     ) -> Optional[models.ChatCompletionResponse]:
-        r"""Chat Completion
+        r"""Agents Completion
 
         :param messages: The prompt(s) to generate completions for, encoded as a list of dict with role and content.
         :param agent_id: The ID of the agent to use for this completion.
@@ -221,16 +221,16 @@ class Agents(BaseSDK):
     
     def stream(
         self, *,
-        model: Nullable[str],
-        prompt: str,
-        temperature: Optional[float] = 0.7,
-        top_p: Optional[float] = 1,
+        messages: Union[List[models.AgentsCompletionStreamRequestMessages], List[models.AgentsCompletionStreamRequestMessagesTypedDict]],
+        agent_id: str,
         max_tokens: OptionalNullable[int] = UNSET,
         min_tokens: OptionalNullable[int] = UNSET,
         stream: Optional[bool] = True,
         stop: Optional[Union[models.AgentsCompletionStreamRequestStop, models.AgentsCompletionStreamRequestStopTypedDict]] = None,
         random_seed: OptionalNullable[int] = UNSET,
-        suffix: OptionalNullable[str] = UNSET,
+        response_format: Optional[Union[models.ResponseFormat, models.ResponseFormatTypedDict]] = None,
+        tools: OptionalNullable[Union[List[models.Tool], List[models.ToolTypedDict]]] = UNSET,
+        tool_choice: Optional[models.AgentsCompletionStreamRequestToolChoice] = "auto",
         retries: OptionalNullable[utils.RetryConfig] = UNSET,
         server_url: Optional[str] = None,
         timeout_ms: Optional[int] = None,
@@ -239,16 +239,16 @@ class Agents(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 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 messages: The prompt(s) to generate completions for, encoded as a list of dict with role and content.
+        :param agent_id: The ID of the agent to use for this completion.
         :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 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 suffix: Optional text/code that adds more context for the model. When given a `prompt` and a `suffix` the model will fill what is between them. When `suffix` is not provided, the model will simply execute completion starting with `prompt`.
+        :param response_format: 
+        :param tools: 
+        :param tool_choice: 
         :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
@@ -262,16 +262,16 @@ class Agents(BaseSDK):
             base_url = server_url
         
         request = models.AgentsCompletionStreamRequest(
-            model=model,
-            temperature=temperature,
-            top_p=top_p,
             max_tokens=max_tokens,
             min_tokens=min_tokens,
             stream=stream,
             stop=stop,
             random_seed=random_seed,
-            prompt=prompt,
-            suffix=suffix,
+            messages=utils.get_pydantic_model(messages, List[models.AgentsCompletionStreamRequestMessages]),
+            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,
+            agent_id=agent_id,
         )
         
         req = self.build_request(
@@ -328,16 +328,16 @@ class Agents(BaseSDK):
     
     async def stream_async(
         self, *,
-        model: Nullable[str],
-        prompt: str,
-        temperature: Optional[float] = 0.7,
-        top_p: Optional[float] = 1,
+        messages: Union[List[models.AgentsCompletionStreamRequestMessages], List[models.AgentsCompletionStreamRequestMessagesTypedDict]],
+        agent_id: str,
         max_tokens: OptionalNullable[int] = UNSET,
         min_tokens: OptionalNullable[int] = UNSET,
         stream: Optional[bool] = True,
         stop: Optional[Union[models.AgentsCompletionStreamRequestStop, models.AgentsCompletionStreamRequestStopTypedDict]] = None,
         random_seed: OptionalNullable[int] = UNSET,
-        suffix: OptionalNullable[str] = UNSET,
+        response_format: Optional[Union[models.ResponseFormat, models.ResponseFormatTypedDict]] = None,
+        tools: OptionalNullable[Union[List[models.Tool], List[models.ToolTypedDict]]] = UNSET,
+        tool_choice: Optional[models.AgentsCompletionStreamRequestToolChoice] = "auto",
         retries: OptionalNullable[utils.RetryConfig] = UNSET,
         server_url: Optional[str] = None,
         timeout_ms: Optional[int] = None,
@@ -346,16 +346,16 @@ class Agents(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 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 messages: The prompt(s) to generate completions for, encoded as a list of dict with role and content.
+        :param agent_id: The ID of the agent to use for this completion.
         :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 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 suffix: Optional text/code that adds more context for the model. When given a `prompt` and a `suffix` the model will fill what is between them. When `suffix` is not provided, the model will simply execute completion starting with `prompt`.
+        :param response_format: 
+        :param tools: 
+        :param tool_choice: 
         :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
@@ -369,16 +369,16 @@ class Agents(BaseSDK):
             base_url = server_url
         
         request = models.AgentsCompletionStreamRequest(
-            model=model,
-            temperature=temperature,
-            top_p=top_p,
             max_tokens=max_tokens,
             min_tokens=min_tokens,
             stream=stream,
             stop=stop,
             random_seed=random_seed,
-            prompt=prompt,
-            suffix=suffix,
+            messages=utils.get_pydantic_model(messages, List[models.AgentsCompletionStreamRequestMessages]),
+            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,
+            agent_id=agent_id,
         )
         
         req = self.build_request(

mistralai/chat.py

@@ -32,7 +32,7 @@ class Chat(BaseSDK):
     ) -> Optional[models.ChatCompletionResponse]:
         r"""Chat Completion
 
-        :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 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.
         :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.
@@ -145,7 +145,7 @@ class Chat(BaseSDK):
     ) -> Optional[models.ChatCompletionResponse]:
         r"""Chat Completion
 
-        :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 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.
         :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.
@@ -260,7 +260,7 @@ 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 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 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.
         :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.
@@ -376,7 +376,7 @@ 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 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 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.
         :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.

mistralai/client.py

@@ -1,6 +1,6 @@
 from typing import Optional
 
-MIGRATION_MESSAGE = "This client is deprecated starting v1.0.0, pin your version to 0.4.2. Or migrate by following this documentation: TODO add link"
+MIGRATION_MESSAGE = "This client is deprecated. To migrate to the new client, please refer to this guide: https://github.com/mistralai/client-python/blob/main/MIGRATION.md. If you need to use this client anyway, pin your version to 0.4.2."
 
 
 class MistralClient:

mistralai/jobs.py

@@ -211,12 +211,11 @@ class Jobs(BaseSDK):
         self, *,
         model: models.FineTuneableModel,
         hyperparameters: Union[models.TrainingParametersIn, models.TrainingParametersInTypedDict],
-        dry_run: OptionalNullable[bool] = UNSET,
         training_files: Optional[Union[List[models.TrainingFile], List[models.TrainingFileTypedDict]]] = None,
         validation_files: OptionalNullable[List[str]] = UNSET,
         suffix: OptionalNullable[str] = UNSET,
-        integrations: OptionalNullable[Union[List[models.WandbIntegration], List[models.WandbIntegrationTypedDict]]] = UNSET,
-        repositories: Optional[Union[List[models.GithubRepositoryIn], List[models.GithubRepositoryInTypedDict]]] = None,
+        integrations: OptionalNullable[Union[List[models.JobInIntegrations], List[models.JobInIntegrationsTypedDict]]] = UNSET,
+        repositories: Optional[Union[List[models.JobInRepositories], List[models.JobInRepositoriesTypedDict]]] = None,
         auto_start: Optional[bool] = None,
         retries: OptionalNullable[utils.RetryConfig] = UNSET,
         server_url: Optional[str] = None,
@@ -228,7 +227,6 @@ class Jobs(BaseSDK):
 
         :param model: The name of the model to fine-tune.
         :param hyperparameters: The fine-tuning hyperparameter settings used in a fine-tune job.
-        :param dry_run: * If `true` the job is not spawned, instead the query returns a handful of useful metadata   for the user to perform sanity checks (see `LegacyJobMetadataOut` response). * Otherwise, the job is started and the query returns the job ID along with some of the   input parameters (see `JobOut` response).
         :param training_files: 
         :param validation_files: A list containing the IDs of uploaded files that contain validation data. If you provide these files, the data is used to generate validation metrics periodically during fine-tuning. These metrics can be viewed in `checkpoints` when getting the status of a running fine-tuning job. The same data should not be present in both train and validation files.
         :param suffix: A string that will be added to your fine-tuning model name. For example, a suffix of \"my-great-model\" would produce a model name like `ft:open-mistral-7b:my-great-model:xxx...`
@@ -247,18 +245,15 @@ class Jobs(BaseSDK):
         if server_url is not None:
             base_url = server_url
         
-        request = models.JobsAPIRoutesFineTuningCreateFineTuningJobRequest(
-            dry_run=dry_run,
-            job_in=models.JobIn(
-                model=model,
-                training_files=utils.get_pydantic_model(training_files, Optional[List[models.TrainingFile]]),
-                validation_files=validation_files,
-                hyperparameters=utils.get_pydantic_model(hyperparameters, models.TrainingParametersIn),
-                suffix=suffix,
-                integrations=utils.get_pydantic_model(integrations, OptionalNullable[List[models.WandbIntegration]]),
-                repositories=utils.get_pydantic_model(repositories, Optional[List[models.GithubRepositoryIn]]),
-                auto_start=auto_start,
-            ),
+        request = models.JobIn(
+            model=model,
+            training_files=utils.get_pydantic_model(training_files, Optional[List[models.TrainingFile]]),
+            validation_files=validation_files,
+            hyperparameters=utils.get_pydantic_model(hyperparameters, models.TrainingParametersIn),
+            suffix=suffix,
+            integrations=utils.get_pydantic_model(integrations, OptionalNullable[List[models.JobInIntegrations]]),
+            repositories=utils.get_pydantic_model(repositories, Optional[List[models.JobInRepositories]]),
+            auto_start=auto_start,
         )
         
         req = self.build_request(
@@ -273,7 +268,7 @@ class Jobs(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.job_in, False, False, "json", models.JobIn),
+            get_serialized_body=lambda: utils.serialize_request_body(request, False, False, "json", models.JobIn),
             timeout_ms=timeout_ms,
         )
         
@@ -312,12 +307,11 @@ class Jobs(BaseSDK):
         self, *,
         model: models.FineTuneableModel,
         hyperparameters: Union[models.TrainingParametersIn, models.TrainingParametersInTypedDict],
-        dry_run: OptionalNullable[bool] = UNSET,
         training_files: Optional[Union[List[models.TrainingFile], List[models.TrainingFileTypedDict]]] = None,
         validation_files: OptionalNullable[List[str]] = UNSET,
         suffix: OptionalNullable[str] = UNSET,
-        integrations: OptionalNullable[Union[List[models.WandbIntegration], List[models.WandbIntegrationTypedDict]]] = UNSET,
-        repositories: Optional[Union[List[models.GithubRepositoryIn], List[models.GithubRepositoryInTypedDict]]] = None,
+        integrations: OptionalNullable[Union[List[models.JobInIntegrations], List[models.JobInIntegrationsTypedDict]]] = UNSET,
+        repositories: Optional[Union[List[models.JobInRepositories], List[models.JobInRepositoriesTypedDict]]] = None,
         auto_start: Optional[bool] = None,
         retries: OptionalNullable[utils.RetryConfig] = UNSET,
         server_url: Optional[str] = None,
@@ -329,7 +323,6 @@ class Jobs(BaseSDK):
 
         :param model: The name of the model to fine-tune.
         :param hyperparameters: The fine-tuning hyperparameter settings used in a fine-tune job.
-        :param dry_run: * If `true` the job is not spawned, instead the query returns a handful of useful metadata   for the user to perform sanity checks (see `LegacyJobMetadataOut` response). * Otherwise, the job is started and the query returns the job ID along with some of the   input parameters (see `JobOut` response).
         :param training_files: 
         :param validation_files: A list containing the IDs of uploaded files that contain validation data. If you provide these files, the data is used to generate validation metrics periodically during fine-tuning. These metrics can be viewed in `checkpoints` when getting the status of a running fine-tuning job. The same data should not be present in both train and validation files.
         :param suffix: A string that will be added to your fine-tuning model name. For example, a suffix of \"my-great-model\" would produce a model name like `ft:open-mistral-7b:my-great-model:xxx...`
@@ -348,18 +341,15 @@ class Jobs(BaseSDK):
         if server_url is not None:
             base_url = server_url
         
-        request = models.JobsAPIRoutesFineTuningCreateFineTuningJobRequest(
-            dry_run=dry_run,
-            job_in=models.JobIn(
-                model=model,
-                training_files=utils.get_pydantic_model(training_files, Optional[List[models.TrainingFile]]),
-                validation_files=validation_files,
-                hyperparameters=utils.get_pydantic_model(hyperparameters, models.TrainingParametersIn),
-                suffix=suffix,
-                integrations=utils.get_pydantic_model(integrations, OptionalNullable[List[models.WandbIntegration]]),
-                repositories=utils.get_pydantic_model(repositories, Optional[List[models.GithubRepositoryIn]]),
-                auto_start=auto_start,
-            ),
+        request = models.JobIn(
+            model=model,
+            training_files=utils.get_pydantic_model(training_files, Optional[List[models.TrainingFile]]),
+            validation_files=validation_files,
+            hyperparameters=utils.get_pydantic_model(hyperparameters, models.TrainingParametersIn),
+            suffix=suffix,
+            integrations=utils.get_pydantic_model(integrations, OptionalNullable[List[models.JobInIntegrations]]),
+            repositories=utils.get_pydantic_model(repositories, Optional[List[models.JobInRepositories]]),
+            auto_start=auto_start,
         )
         
         req = self.build_request(
@@ -374,7 +364,7 @@ class Jobs(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.job_in, False, False, "json", models.JobIn),
+            get_serialized_body=lambda: utils.serialize_request_body(request, False, False, "json", models.JobIn),
             timeout_ms=timeout_ms,
         )
         

mistralai/models/__init__.py

@@ -1,8 +1,8 @@
 """Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT."""
 
 from .agentscompletionrequest import AgentsCompletionRequest, AgentsCompletionRequestMessages, AgentsCompletionRequestMessagesTypedDict, AgentsCompletionRequestStop, AgentsCompletionRequestStopTypedDict, AgentsCompletionRequestToolChoice, AgentsCompletionRequestTypedDict
-from .agentscompletionstreamrequest import AgentsCompletionStreamRequest, AgentsCompletionStreamRequestStop, AgentsCompletionStreamRequestStopTypedDict, AgentsCompletionStreamRequestTypedDict
-from .archiveftmodelout import ArchiveFTModelOut, ArchiveFTModelOutTypedDict
+from .agentscompletionstreamrequest import AgentsCompletionStreamRequest, AgentsCompletionStreamRequestMessages, AgentsCompletionStreamRequestMessagesTypedDict, AgentsCompletionStreamRequestStop, AgentsCompletionStreamRequestStopTypedDict, AgentsCompletionStreamRequestToolChoice, AgentsCompletionStreamRequestTypedDict
+from .archiveftmodelout import ArchiveFTModelOut, ArchiveFTModelOutObject, ArchiveFTModelOutTypedDict
 from .assistantmessage import AssistantMessage, AssistantMessageRole, AssistantMessageTypedDict
 from .chatcompletionchoice import ChatCompletionChoice, ChatCompletionChoiceTypedDict, FinishReason
 from .chatcompletionrequest import ChatCompletionRequest, ChatCompletionRequestTypedDict, Messages, MessagesTypedDict, Stop, StopTypedDict, ToolChoice
@@ -17,39 +17,39 @@ from .delete_model_v1_models_model_id_deleteop import DeleteModelV1ModelsModelID
 from .deletefileout import DeleteFileOut, DeleteFileOutTypedDict
 from .deletemodelout import DeleteModelOut, DeleteModelOutTypedDict
 from .deltamessage import DeltaMessage, DeltaMessageTypedDict
-from .detailedjobout import DetailedJobOut, DetailedJobOutStatus, DetailedJobOutTypedDict
+from .detailedjobout import DetailedJobOut, DetailedJobOutIntegrations, DetailedJobOutIntegrationsTypedDict, DetailedJobOutObject, DetailedJobOutRepositories, DetailedJobOutRepositoriesTypedDict, DetailedJobOutStatus, DetailedJobOutTypedDict
 from .embeddingrequest import EmbeddingRequest, EmbeddingRequestTypedDict, Inputs, InputsTypedDict
 from .embeddingresponse import EmbeddingResponse, EmbeddingResponseTypedDict
 from .embeddingresponsedata import EmbeddingResponseData, EmbeddingResponseDataTypedDict
 from .eventout import EventOut, EventOutTypedDict
 from .files_api_routes_delete_fileop import FilesAPIRoutesDeleteFileRequest, FilesAPIRoutesDeleteFileRequestTypedDict
 from .files_api_routes_retrieve_fileop import FilesAPIRoutesRetrieveFileRequest, FilesAPIRoutesRetrieveFileRequestTypedDict
-from .files_api_routes_upload_fileop import File, FileTypedDict, FilesAPIRoutesUploadFileMultiPartBodyParams, FilesAPIRoutesUploadFileMultiPartBodyParamsTypedDict
-from .fileschema import FileSchema, FileSchemaTypedDict
+from .files_api_routes_upload_fileop import File, FileTypedDict, FilesAPIRoutesUploadFileMultiPartBodyParams, FilesAPIRoutesUploadFileMultiPartBodyParamsTypedDict, FilesAPIRoutesUploadFilePurpose
+from .fileschema import FileSchema, FileSchemaPurpose, FileSchemaTypedDict
 from .fimcompletionrequest import FIMCompletionRequest, FIMCompletionRequestStop, FIMCompletionRequestStopTypedDict, FIMCompletionRequestTypedDict
 from .fimcompletionresponse import FIMCompletionResponse, FIMCompletionResponseTypedDict
 from .fimcompletionstreamrequest import FIMCompletionStreamRequest, FIMCompletionStreamRequestStop, FIMCompletionStreamRequestStopTypedDict, FIMCompletionStreamRequestTypedDict
 from .finetuneablemodel import FineTuneableModel
 from .ftmodelcapabilitiesout import FTModelCapabilitiesOut, FTModelCapabilitiesOutTypedDict
-from .ftmodelout import FTModelOut, FTModelOutTypedDict
+from .ftmodelout import FTModelOut, FTModelOutObject, FTModelOutTypedDict
 from .function import Function, FunctionTypedDict
-from .functioncall import FunctionCall, FunctionCallTypedDict
-from .githubrepositoryin import GithubRepositoryIn, GithubRepositoryInTypedDict
-from .githubrepositoryout import GithubRepositoryOut, GithubRepositoryOutTypedDict
+from .functioncall import Arguments, ArgumentsTypedDict, FunctionCall, FunctionCallTypedDict
+from .githubrepositoryin import GithubRepositoryIn, GithubRepositoryInType, GithubRepositoryInTypedDict
+from .githubrepositoryout import GithubRepositoryOut, GithubRepositoryOutType, GithubRepositoryOutTypedDict
 from .httpvalidationerror import HTTPValidationError, HTTPValidationErrorData
-from .jobin import JobIn, JobInTypedDict
+from .jobin import JobIn, JobInIntegrations, JobInIntegrationsTypedDict, JobInRepositories, JobInRepositoriesTypedDict, JobInTypedDict
 from .jobmetadataout import JobMetadataOut, JobMetadataOutTypedDict
-from .jobout import JobOut, JobOutTypedDict, Status
+from .jobout import Integrations, IntegrationsTypedDict, JobOut, JobOutTypedDict, Object, Repositories, RepositoriesTypedDict, Status
 from .jobs_api_routes_fine_tuning_archive_fine_tuned_modelop import JobsAPIRoutesFineTuningArchiveFineTunedModelRequest, JobsAPIRoutesFineTuningArchiveFineTunedModelRequestTypedDict
 from .jobs_api_routes_fine_tuning_cancel_fine_tuning_jobop import JobsAPIRoutesFineTuningCancelFineTuningJobRequest, JobsAPIRoutesFineTuningCancelFineTuningJobRequestTypedDict
-from .jobs_api_routes_fine_tuning_create_fine_tuning_jobop import JobsAPIRoutesFineTuningCreateFineTuningJobRequest, JobsAPIRoutesFineTuningCreateFineTuningJobRequestTypedDict, JobsAPIRoutesFineTuningCreateFineTuningJobResponse, JobsAPIRoutesFineTuningCreateFineTuningJobResponseTypedDict
+from .jobs_api_routes_fine_tuning_create_fine_tuning_jobop import JobsAPIRoutesFineTuningCreateFineTuningJobResponse, JobsAPIRoutesFineTuningCreateFineTuningJobResponseTypedDict
 from .jobs_api_routes_fine_tuning_get_fine_tuning_jobop import JobsAPIRoutesFineTuningGetFineTuningJobRequest, JobsAPIRoutesFineTuningGetFineTuningJobRequestTypedDict
 from .jobs_api_routes_fine_tuning_get_fine_tuning_jobsop import JobsAPIRoutesFineTuningGetFineTuningJobsRequest, JobsAPIRoutesFineTuningGetFineTuningJobsRequestTypedDict, QueryParamStatus
 from .jobs_api_routes_fine_tuning_start_fine_tuning_jobop import JobsAPIRoutesFineTuningStartFineTuningJobRequest, JobsAPIRoutesFineTuningStartFineTuningJobRequestTypedDict
 from .jobs_api_routes_fine_tuning_unarchive_fine_tuned_modelop import JobsAPIRoutesFineTuningUnarchiveFineTunedModelRequest, JobsAPIRoutesFineTuningUnarchiveFineTunedModelRequestTypedDict
 from .jobs_api_routes_fine_tuning_update_fine_tuned_modelop import JobsAPIRoutesFineTuningUpdateFineTunedModelRequest, JobsAPIRoutesFineTuningUpdateFineTunedModelRequestTypedDict
-from .jobsout import JobsOut, JobsOutTypedDict
-from .legacyjobmetadataout import LegacyJobMetadataOut, LegacyJobMetadataOutTypedDict
+from .jobsout import JobsOut, JobsOutObject, JobsOutTypedDict
+from .legacyjobmetadataout import LegacyJobMetadataOut, LegacyJobMetadataOutObject, LegacyJobMetadataOutTypedDict
 from .listfilesout import ListFilesOut, ListFilesOutTypedDict
 from .metricout import MetricOut, MetricOutTypedDict
 from .modelcapabilities import ModelCapabilities, ModelCapabilitiesTypedDict
@@ -57,26 +57,26 @@ from .modelcard import ModelCard, ModelCardTypedDict
 from .modellist import ModelList, ModelListTypedDict
 from .responseformat import ResponseFormat, ResponseFormatTypedDict, ResponseFormats
 from .retrieve_model_v1_models_model_id_getop import RetrieveModelV1ModelsModelIDGetRequest, RetrieveModelV1ModelsModelIDGetRequestTypedDict
-from .retrievefileout import RetrieveFileOut, RetrieveFileOutTypedDict
+from .retrievefileout import RetrieveFileOut, RetrieveFileOutPurpose, RetrieveFileOutTypedDict
 from .sampletype import SampleType
 from .sdkerror import SDKError
 from .security import Security, SecurityTypedDict
 from .source import Source
 from .systemmessage import Content, ContentTypedDict, Role, SystemMessage, SystemMessageTypedDict
 from .textchunk import TextChunk, TextChunkTypedDict
-from .tool import Tool, ToolTypedDict
-from .toolcall import ToolCall, ToolCallTypedDict
+from .tool import Tool, ToolToolTypes, ToolTypedDict
+from .toolcall import ToolCall, ToolCallTypedDict, ToolTypes
 from .toolmessage import ToolMessage, ToolMessageRole, ToolMessageTypedDict
 from .trainingfile import TrainingFile, TrainingFileTypedDict
 from .trainingparameters import TrainingParameters, TrainingParametersTypedDict
 from .trainingparametersin import TrainingParametersIn, TrainingParametersInTypedDict
-from .unarchiveftmodelout import UnarchiveFTModelOut, UnarchiveFTModelOutTypedDict
+from .unarchiveftmodelout import UnarchiveFTModelOut, UnarchiveFTModelOutObject, UnarchiveFTModelOutTypedDict
 from .updateftmodelin import UpdateFTModelIn, UpdateFTModelInTypedDict
-from .uploadfileout import UploadFileOut, UploadFileOutTypedDict
+from .uploadfileout import Purpose, UploadFileOut, UploadFileOutTypedDict
 from .usageinfo import UsageInfo, UsageInfoTypedDict
 from .usermessage import UserMessage, UserMessageContent, UserMessageContentTypedDict, UserMessageRole, UserMessageTypedDict
 from .validationerror import Loc, LocTypedDict, ValidationError, ValidationErrorTypedDict
-from .wandbintegration import WandbIntegration, WandbIntegrationTypedDict
-from .wandbintegrationout import WandbIntegrationOut, WandbIntegrationOutTypedDict
+from .wandbintegration import WandbIntegration, WandbIntegrationType, WandbIntegrationTypedDict
+from .wandbintegrationout import Type, WandbIntegrationOut, WandbIntegrationOutTypedDict
 
-__all__ = ["AgentsCompletionRequest", "AgentsCompletionRequestMessages", "AgentsCompletionRequestMessagesTypedDict", "AgentsCompletionRequestStop", "AgentsCompletionRequestStopTypedDict", "AgentsCompletionRequestToolChoice", "AgentsCompletionRequestTypedDict", "AgentsCompletionStreamRequest", "AgentsCompletionStreamRequestStop", "AgentsCompletionStreamRequestStopTypedDict", "AgentsCompletionStreamRequestTypedDict", "ArchiveFTModelOut", "ArchiveFTModelOutTypedDict", "AssistantMessage", "AssistantMessageRole", "AssistantMessageTypedDict", "ChatCompletionChoice", "ChatCompletionChoiceTypedDict", "ChatCompletionRequest", "ChatCompletionRequestTypedDict", "ChatCompletionResponse", "ChatCompletionResponseTypedDict", "ChatCompletionStreamRequest", "ChatCompletionStreamRequestMessages", "ChatCompletionStreamRequestMessagesTypedDict", "ChatCompletionStreamRequestStop", "ChatCompletionStreamRequestStopTypedDict", "ChatCompletionStreamRequestToolChoice", "ChatCompletionStreamRequestTypedDict", "CheckpointOut", "CheckpointOutTypedDict", "CompletionChunk", "CompletionChunkTypedDict", "CompletionEvent", "CompletionEventTypedDict", "CompletionResponseStreamChoice", "CompletionResponseStreamChoiceFinishReason", "CompletionResponseStreamChoiceTypedDict", "Content", "ContentChunk", "ContentChunkTypedDict", "ContentTypedDict", "DeleteFileOut", "DeleteFileOutTypedDict", "DeleteModelOut", "DeleteModelOutTypedDict", "DeleteModelV1ModelsModelIDDeleteRequest", "DeleteModelV1ModelsModelIDDeleteRequestTypedDict", "DeltaMessage", "DeltaMessageTypedDict", "DetailedJobOut", "DetailedJobOutStatus", "DetailedJobOutTypedDict", "EmbeddingRequest", "EmbeddingRequestTypedDict", "EmbeddingResponse", "EmbeddingResponseData", "EmbeddingResponseDataTypedDict", "EmbeddingResponseTypedDict", "EventOut", "EventOutTypedDict", "FIMCompletionRequest", "FIMCompletionRequestStop", "FIMCompletionRequestStopTypedDict", "FIMCompletionRequestTypedDict", "FIMCompletionResponse", "FIMCompletionResponseTypedDict", "FIMCompletionStreamRequest", "FIMCompletionStreamRequestStop", "FIMCompletionStreamRequestStopTypedDict", "FIMCompletionStreamRequestTypedDict", "FTModelCapabilitiesOut", "FTModelCapabilitiesOutTypedDict", "FTModelOut", "FTModelOutTypedDict", "File", "FileSchema", "FileSchemaTypedDict", "FileTypedDict", "FilesAPIRoutesDeleteFileRequest", "FilesAPIRoutesDeleteFileRequestTypedDict", "FilesAPIRoutesRetrieveFileRequest", "FilesAPIRoutesRetrieveFileRequestTypedDict", "FilesAPIRoutesUploadFileMultiPartBodyParams", "FilesAPIRoutesUploadFileMultiPartBodyParamsTypedDict", "FineTuneableModel", "FinishReason", "Function", "FunctionCall", "FunctionCallTypedDict", "FunctionTypedDict", "GithubRepositoryIn", "GithubRepositoryInTypedDict", "GithubRepositoryOut", "GithubRepositoryOutTypedDict", "HTTPValidationError", "HTTPValidationErrorData", "Inputs", "InputsTypedDict", "JobIn", "JobInTypedDict", "JobMetadataOut", "JobMetadataOutTypedDict", "JobOut", "JobOutTypedDict", "JobsAPIRoutesFineTuningArchiveFineTunedModelRequest", "JobsAPIRoutesFineTuningArchiveFineTunedModelRequestTypedDict", "JobsAPIRoutesFineTuningCancelFineTuningJobRequest", "JobsAPIRoutesFineTuningCancelFineTuningJobRequestTypedDict", "JobsAPIRoutesFineTuningCreateFineTuningJobRequest", "JobsAPIRoutesFineTuningCreateFineTuningJobRequestTypedDict", "JobsAPIRoutesFineTuningCreateFineTuningJobResponse", "JobsAPIRoutesFineTuningCreateFineTuningJobResponseTypedDict", "JobsAPIRoutesFineTuningGetFineTuningJobRequest", "JobsAPIRoutesFineTuningGetFineTuningJobRequestTypedDict", "JobsAPIRoutesFineTuningGetFineTuningJobsRequest", "JobsAPIRoutesFineTuningGetFineTuningJobsRequestTypedDict", "JobsAPIRoutesFineTuningStartFineTuningJobRequest", "JobsAPIRoutesFineTuningStartFineTuningJobRequestTypedDict", "JobsAPIRoutesFineTuningUnarchiveFineTunedModelRequest", "JobsAPIRoutesFineTuningUnarchiveFineTunedModelRequestTypedDict", "JobsAPIRoutesFineTuningUpdateFineTunedModelRequest", "JobsAPIRoutesFineTuningUpdateFineTunedModelRequestTypedDict", "JobsOut", "JobsOutTypedDict", "LegacyJobMetadataOut", "LegacyJobMetadataOutTypedDict", "ListFilesOut", "ListFilesOutTypedDict", "Loc", "LocTypedDict", "Messages", "MessagesTypedDict", "MetricOut", "MetricOutTypedDict", "ModelCapabilities", "ModelCapabilitiesTypedDict", "ModelCard", "ModelCardTypedDict", "ModelList", "ModelListTypedDict", "QueryParamStatus", "ResponseFormat", "ResponseFormatTypedDict", "ResponseFormats", "RetrieveFileOut", "RetrieveFileOutTypedDict", "RetrieveModelV1ModelsModelIDGetRequest", "RetrieveModelV1ModelsModelIDGetRequestTypedDict", "Role", "SDKError", "SampleType", "Security", "SecurityTypedDict", "Source", "Status", "Stop", "StopTypedDict", "SystemMessage", "SystemMessageTypedDict", "TextChunk", "TextChunkTypedDict", "Tool", "ToolCall", "ToolCallTypedDict", "ToolChoice", "ToolMessage", "ToolMessageRole", "ToolMessageTypedDict", "ToolTypedDict", "TrainingFile", "TrainingFileTypedDict", "TrainingParameters", "TrainingParametersIn", "TrainingParametersInTypedDict", "TrainingParametersTypedDict", "UnarchiveFTModelOut", "UnarchiveFTModelOutTypedDict", "UpdateFTModelIn", "UpdateFTModelInTypedDict", "UploadFileOut", "UploadFileOutTypedDict", "UsageInfo", "UsageInfoTypedDict", "UserMessage", "UserMessageContent", "UserMessageContentTypedDict", "UserMessageRole", "UserMessageTypedDict", "ValidationError", "ValidationErrorTypedDict", "WandbIntegration", "WandbIntegrationOut", "WandbIntegrationOutTypedDict", "WandbIntegrationTypedDict"]
+__all__ = ["AgentsCompletionRequest", "AgentsCompletionRequestMessages", "AgentsCompletionRequestMessagesTypedDict", "AgentsCompletionRequestStop", "AgentsCompletionRequestStopTypedDict", "AgentsCompletionRequestToolChoice", "AgentsCompletionRequestTypedDict", "AgentsCompletionStreamRequest", "AgentsCompletionStreamRequestMessages", "AgentsCompletionStreamRequestMessagesTypedDict", "AgentsCompletionStreamRequestStop", "AgentsCompletionStreamRequestStopTypedDict", "AgentsCompletionStreamRequestToolChoice", "AgentsCompletionStreamRequestTypedDict", "ArchiveFTModelOut", "ArchiveFTModelOutObject", "ArchiveFTModelOutTypedDict", "Arguments", "ArgumentsTypedDict", "AssistantMessage", "AssistantMessageRole", "AssistantMessageTypedDict", "ChatCompletionChoice", "ChatCompletionChoiceTypedDict", "ChatCompletionRequest", "ChatCompletionRequestTypedDict", "ChatCompletionResponse", "ChatCompletionResponseTypedDict", "ChatCompletionStreamRequest", "ChatCompletionStreamRequestMessages", "ChatCompletionStreamRequestMessagesTypedDict", "ChatCompletionStreamRequestStop", "ChatCompletionStreamRequestStopTypedDict", "ChatCompletionStreamRequestToolChoice", "ChatCompletionStreamRequestTypedDict", "CheckpointOut", "CheckpointOutTypedDict", "CompletionChunk", "CompletionChunkTypedDict", "CompletionEvent", "CompletionEventTypedDict", "CompletionResponseStreamChoice", "CompletionResponseStreamChoiceFinishReason", "CompletionResponseStreamChoiceTypedDict", "Content", "ContentChunk", "ContentChunkTypedDict", "ContentTypedDict", "DeleteFileOut", "DeleteFileOutTypedDict", "DeleteModelOut", "DeleteModelOutTypedDict", "DeleteModelV1ModelsModelIDDeleteRequest", "DeleteModelV1ModelsModelIDDeleteRequestTypedDict", "DeltaMessage", "DeltaMessageTypedDict", "DetailedJobOut", "DetailedJobOutIntegrations", "DetailedJobOutIntegrationsTypedDict", "DetailedJobOutObject", "DetailedJobOutRepositories", "DetailedJobOutRepositoriesTypedDict", "DetailedJobOutStatus", "DetailedJobOutTypedDict", "EmbeddingRequest", "EmbeddingRequestTypedDict", "EmbeddingResponse", "EmbeddingResponseData", "EmbeddingResponseDataTypedDict", "EmbeddingResponseTypedDict", "EventOut", "EventOutTypedDict", "FIMCompletionRequest", "FIMCompletionRequestStop", "FIMCompletionRequestStopTypedDict", "FIMCompletionRequestTypedDict", "FIMCompletionResponse", "FIMCompletionResponseTypedDict", "FIMCompletionStreamRequest", "FIMCompletionStreamRequestStop", "FIMCompletionStreamRequestStopTypedDict", "FIMCompletionStreamRequestTypedDict", "FTModelCapabilitiesOut", "FTModelCapabilitiesOutTypedDict", "FTModelOut", "FTModelOutObject", "FTModelOutTypedDict", "File", "FileSchema", "FileSchemaPurpose", "FileSchemaTypedDict", "FileTypedDict", "FilesAPIRoutesDeleteFileRequest", "FilesAPIRoutesDeleteFileRequestTypedDict", "FilesAPIRoutesRetrieveFileRequest", "FilesAPIRoutesRetrieveFileRequestTypedDict", "FilesAPIRoutesUploadFileMultiPartBodyParams", "FilesAPIRoutesUploadFileMultiPartBodyParamsTypedDict", "FilesAPIRoutesUploadFilePurpose", "FineTuneableModel", "FinishReason", "Function", "FunctionCall", "FunctionCallTypedDict", "FunctionTypedDict", "GithubRepositoryIn", "GithubRepositoryInType", "GithubRepositoryInTypedDict", "GithubRepositoryOut", "GithubRepositoryOutType", "GithubRepositoryOutTypedDict", "HTTPValidationError", "HTTPValidationErrorData", "Inputs", "InputsTypedDict", "Integrations", "IntegrationsTypedDict", "JobIn", "JobInIntegrations", "JobInIntegrationsTypedDict", "JobInRepositories", "JobInRepositoriesTypedDict", "JobInTypedDict", "JobMetadataOut", "JobMetadataOutTypedDict", "JobOut", "JobOutTypedDict", "JobsAPIRoutesFineTuningArchiveFineTunedModelRequest", "JobsAPIRoutesFineTuningArchiveFineTunedModelRequestTypedDict", "JobsAPIRoutesFineTuningCancelFineTuningJobRequest", "JobsAPIRoutesFineTuningCancelFineTuningJobRequestTypedDict", "JobsAPIRoutesFineTuningCreateFineTuningJobResponse", "JobsAPIRoutesFineTuningCreateFineTuningJobResponseTypedDict", "JobsAPIRoutesFineTuningGetFineTuningJobRequest", "JobsAPIRoutesFineTuningGetFineTuningJobRequestTypedDict", "JobsAPIRoutesFineTuningGetFineTuningJobsRequest", "JobsAPIRoutesFineTuningGetFineTuningJobsRequestTypedDict", "JobsAPIRoutesFineTuningStartFineTuningJobRequest", "JobsAPIRoutesFineTuningStartFineTuningJobRequestTypedDict", "JobsAPIRoutesFineTuningUnarchiveFineTunedModelRequest", "JobsAPIRoutesFineTuningUnarchiveFineTunedModelRequestTypedDict", "JobsAPIRoutesFineTuningUpdateFineTunedModelRequest", "JobsAPIRoutesFineTuningUpdateFineTunedModelRequestTypedDict", "JobsOut", "JobsOutObject", "JobsOutTypedDict", "LegacyJobMetadataOut", "LegacyJobMetadataOutObject", "LegacyJobMetadataOutTypedDict", "ListFilesOut", "ListFilesOutTypedDict", "Loc", "LocTypedDict", "Messages", "MessagesTypedDict", "MetricOut", "MetricOutTypedDict", "ModelCapabilities", "ModelCapabilitiesTypedDict", "ModelCard", "ModelCardTypedDict", "ModelList", "ModelListTypedDict", "Object", "Purpose", "QueryParamStatus", "Repositories", "RepositoriesTypedDict", "ResponseFormat", "ResponseFormatTypedDict", "ResponseFormats", "RetrieveFileOut", "RetrieveFileOutPurpose", "RetrieveFileOutTypedDict", "RetrieveModelV1ModelsModelIDGetRequest", "RetrieveModelV1ModelsModelIDGetRequestTypedDict", "Role", "SDKError", "SampleType", "Security", "SecurityTypedDict", "Source", "Status", "Stop", "StopTypedDict", "SystemMessage", "SystemMessageTypedDict", "TextChunk", "TextChunkTypedDict", "Tool", "ToolCall", "ToolCallTypedDict", "ToolChoice", "ToolMessage", "ToolMessageRole", "ToolMessageTypedDict", "ToolToolTypes", "ToolTypedDict", "ToolTypes", "TrainingFile", "TrainingFileTypedDict", "TrainingParameters", "TrainingParametersIn", "TrainingParametersInTypedDict", "TrainingParametersTypedDict", "Type", "UnarchiveFTModelOut", "UnarchiveFTModelOutObject", "UnarchiveFTModelOutTypedDict", "UpdateFTModelIn", "UpdateFTModelInTypedDict", "UploadFileOut", "UploadFileOutTypedDict", "UsageInfo", "UsageInfoTypedDict", "UserMessage", "UserMessageContent", "UserMessageContentTypedDict", "UserMessageRole", "UserMessageTypedDict", "ValidationError", "ValidationErrorTypedDict", "WandbIntegration", "WandbIntegrationOut", "WandbIntegrationOutTypedDict", "WandbIntegrationType", "WandbIntegrationTypedDict"]

mistralai/models/agentscompletionrequest.py

@@ -13,6 +13,20 @@ from typing import List, Literal, Optional, TypedDict, Union
 from typing_extensions import Annotated, NotRequired
 
 
+AgentsCompletionRequestStopTypedDict = Union[str, List[str]]
+r"""Stop generation if this token is detected. Or if one of these tokens is detected when providing an array"""
+
+
+AgentsCompletionRequestStop = Union[str, List[str]]
+r"""Stop generation if this token is detected. Or if one of these tokens is detected when providing an array"""
+
+
+AgentsCompletionRequestMessagesTypedDict = Union[UserMessageTypedDict, AssistantMessageTypedDict, ToolMessageTypedDict]
+
+
+AgentsCompletionRequestMessages = Annotated[Union[Annotated[AssistantMessage, Tag("assistant")], Annotated[ToolMessage, Tag("tool")], Annotated[UserMessage, Tag("user")]], Discriminator(lambda m: get_discriminator(m, "role", "role"))]
+
+
 AgentsCompletionRequestToolChoice = Literal["auto", "none", "any"]
 
 class AgentsCompletionRequestTypedDict(TypedDict):
@@ -80,17 +94,3 @@ class AgentsCompletionRequest(BaseModel):
 
         return m
         
-
-AgentsCompletionRequestStopTypedDict = Union[str, List[str]]
-r"""Stop generation if this token is detected. Or if one of these tokens is detected when providing an array"""
-
-
-AgentsCompletionRequestStop = Union[str, List[str]]
-r"""Stop generation if this token is detected. Or if one of these tokens is detected when providing an array"""
-
-
-AgentsCompletionRequestMessagesTypedDict = Union[UserMessageTypedDict, AssistantMessageTypedDict, ToolMessageTypedDict]
-
-
-AgentsCompletionRequestMessages = Annotated[Union[Annotated[AssistantMessage, Tag("assistant")], Annotated[ToolMessage, Tag("tool")], Annotated[UserMessage, Tag("user")]], Discriminator(lambda m: get_discriminator(m, "role", "role"))]
-

mistralai/models/agentscompletionstreamrequest.py

@@ -1,24 +1,39 @@
 """Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT."""
 
 from __future__ import annotations
+from .assistantmessage import AssistantMessage, AssistantMessageTypedDict
+from .responseformat import ResponseFormat, ResponseFormatTypedDict
+from .tool import Tool, ToolTypedDict
+from .toolmessage import ToolMessage, ToolMessageTypedDict
+from .usermessage import UserMessage, UserMessageTypedDict
 from mistralai.types import BaseModel, Nullable, OptionalNullable, UNSET, UNSET_SENTINEL
-from pydantic import model_serializer
-from typing import List, Optional, TypedDict, Union
-from typing_extensions import NotRequired
+from mistralai.utils import get_discriminator
+from pydantic import Discriminator, Tag, model_serializer
+from typing import List, Literal, Optional, TypedDict, Union
+from typing_extensions import Annotated, NotRequired
 
 
+AgentsCompletionStreamRequestStopTypedDict = Union[str, List[str]]
+r"""Stop generation if this token is detected. Or if one of these tokens is detected when providing an array"""
+
+
+AgentsCompletionStreamRequestStop = Union[str, List[str]]
+r"""Stop generation if this token is detected. Or if one of these tokens is detected when providing an array"""
+
+
+AgentsCompletionStreamRequestMessagesTypedDict = Union[UserMessageTypedDict, AssistantMessageTypedDict, ToolMessageTypedDict]
+
+
+AgentsCompletionStreamRequestMessages = Annotated[Union[Annotated[AssistantMessage, Tag("assistant")], Annotated[ToolMessage, Tag("tool")], Annotated[UserMessage, Tag("user")]], Discriminator(lambda m: get_discriminator(m, "role", "role"))]
+
+
+AgentsCompletionStreamRequestToolChoice = Literal["auto", "none", "any"]
+
 class AgentsCompletionStreamRequestTypedDict(TypedDict):
-    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]
-    r"""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."""
+    messages: List[AgentsCompletionStreamRequestMessagesTypedDict]
+    r"""The prompt(s) to generate completions for, encoded as a list of dict with role and content."""
+    agent_id: str
+    r"""The ID of the agent to use for this completion."""
     max_tokens: NotRequired[Nullable[int]]
     r"""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."""
     min_tokens: NotRequired[Nullable[int]]
@@ -28,22 +43,16 @@ class AgentsCompletionStreamRequestTypedDict(TypedDict):
     r"""Stop generation if this token is detected. Or if one of these tokens is detected when providing an array"""
     random_seed: NotRequired[Nullable[int]]
     r"""The seed to use for random sampling. If set, different calls will generate deterministic results."""
-    suffix: NotRequired[Nullable[str]]
-    r"""Optional text/code that adds more context for the model. When given a `prompt` and a `suffix` the model will fill what is between them. When `suffix` is not provided, the model will simply execute completion starting with `prompt`."""
+    response_format: NotRequired[ResponseFormatTypedDict]
+    tools: NotRequired[Nullable[List[ToolTypedDict]]]
+    tool_choice: NotRequired[AgentsCompletionStreamRequestToolChoice]
     
 
 class AgentsCompletionStreamRequest(BaseModel):
-    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
-    r"""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."""
+    messages: List[AgentsCompletionStreamRequestMessages]
+    r"""The prompt(s) to generate completions for, encoded as a list of dict with role and content."""
+    agent_id: str
+    r"""The ID of the agent to use for this completion."""
     max_tokens: OptionalNullable[int] = UNSET
     r"""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."""
     min_tokens: OptionalNullable[int] = UNSET
@@ -53,13 +62,14 @@ class AgentsCompletionStreamRequest(BaseModel):
     r"""Stop generation if this token is detected. Or if one of these tokens is detected when providing an array"""
     random_seed: OptionalNullable[int] = UNSET
     r"""The seed to use for random sampling. If set, different calls will generate deterministic results."""
-    suffix: OptionalNullable[str] = UNSET
-    r"""Optional text/code that adds more context for the model. When given a `prompt` and a `suffix` the model will fill what is between them. When `suffix` is not provided, the model will simply execute completion starting with `prompt`."""
+    response_format: Optional[ResponseFormat] = None
+    tools: OptionalNullable[List[Tool]] = UNSET
+    tool_choice: Optional[AgentsCompletionStreamRequestToolChoice] = "auto"
     
     @model_serializer(mode="wrap")
     def serialize_model(self, handler):
-        optional_fields = ["temperature", "top_p", "max_tokens", "min_tokens", "stream", "stop", "random_seed", "suffix"]
-        nullable_fields = ["model", "max_tokens", "min_tokens", "random_seed", "suffix"]
+        optional_fields = ["max_tokens", "min_tokens", "stream", "stop", "random_seed", "response_format", "tools", "tool_choice"]
+        nullable_fields = ["max_tokens", "min_tokens", "random_seed", "tools"]
         null_default_fields = []
 
         serialized = handler(self)
@@ -82,11 +92,3 @@ class AgentsCompletionStreamRequest(BaseModel):
 
         return m
         
-
-AgentsCompletionStreamRequestStopTypedDict = Union[str, List[str]]
-r"""Stop generation if this token is detected. Or if one of these tokens is detected when providing an array"""
-
-
-AgentsCompletionStreamRequestStop = Union[str, List[str]]
-r"""Stop generation if this token is detected. Or if one of these tokens is detected when providing an array"""
-