### Building and Installing the SDK with Custom Code

Source: https://github.com/mistralai/client-python/blob/main/src/mistralai/extra/README.md

After adding custom code, build the SDK and install it locally. This command sequence removes the existing distribution, builds a new one, and then installs the newly built wheel file.

```bash
rm -rf dist; uv build; uv pip install --reinstall ~/client-python/dist/mistralai-2.0.0-py3-none-any.whl
```

--------------------------------

### Install Mistral Python Client

Source: https://github.com/mistralai/client-python/blob/main/packages/azure/README.md

Install the Mistral AI Python client using pip or uv.

```bash
pip install mistralai
```

```bash
uv add mistralai
```

--------------------------------

### Install Mistral AI GCP Extras

Source: https://github.com/mistralai/client-python/blob/main/packages/gcp/README-PYPI.md

Install the Mistral AI Python client with the necessary extras for Google Cloud Platform integration.

```bash
pip install mistralai[gcp]
```

--------------------------------

### Install Mistral AI GCP Client

Source: https://github.com/mistralai/client-python/blob/main/packages/gcp/README.md

Install the Mistral AI Python client with the necessary extras for Google Cloud integration.

```bash
pip install mistralai[gcp]
```

--------------------------------

### Install Mistral AI Python SDK with GCP Support

Source: https://github.com/mistralai/client-python/blob/main/README.md

Install the Mistral AI Python SDK along with the necessary dependencies for Google Cloud Platform authentication.

```bash
pip install mistralai
# For GCP authentication support (required):
pip install "mistralai[gcp]"
```

--------------------------------

### Install Mistral AI SDK with pip

Source: https://github.com/mistralai/client-python/blob/main/README.md

Use this command to install the mistralai package using the pip package manager.

```bash
pip install mistralai
```

--------------------------------

### Install Mistral AI SDK with agents extra

Source: https://github.com/mistralai/client-python/blob/main/README.md

Install the mistralai package with the 'agents' extra dependencies to enable agent-related features. Requires Python 3.10+.

```bash
pip install "mistralai[agents]"
```

--------------------------------

### ModerationLLMAction Example Usage

Source: https://github.com/mistralai/client-python/blob/main/packages/gcp/docs/models/moderationllmaction.md

Demonstrates how to import and assign a value to ModerationLLMAction. The 'none' value is used as an example.

```python
from mistralai.gcp.client.models import ModerationLLMAction
value: ModerationLLMAction = "none"
```

--------------------------------

### Azure AI Chat Completion Example

Source: https://github.com/mistralai/client-python/blob/main/README.md

Example of how to perform a chat completion using the Mistral AI SDK with Azure AI. Ensure AZURE_API_KEY, AZURE_ENDPOINT, and AZURE_MODEL environment variables are set.

```python
import os
from mistralai.azure.client import MistralAzure

# The SDK automatically injects api-version as a query parameter
client = MistralAzure(
    api_key=os.environ["AZURE_API_KEY"],
    server_url=os.environ["AZURE_ENDPOINT"],
    api_version="2024-05-01-preview",  # Optional, this is the default
)

res = client.chat.complete(
    model=os.environ["AZURE_MODEL"],
    messages=[
        {
            "role": "user",
            "content": "Hello there!",
        }
    ],
)
print(res.choices[0].message.content)
```

--------------------------------

### Google Cloud Chat Completion Example

Source: https://github.com/mistralai/client-python/blob/main/README.md

Example of how to perform a chat completion using the Mistral AI SDK with Google Cloud Vertex AI. The SDK automatically detects credentials and builds the Vertex AI URL.

```python
import os
from mistralai.gcp.client import MistralGCP

# The SDK auto-detects credentials and builds the Vertex AI URL
client = MistralGCP(
    project_id=os.environ.get("GCP_PROJECT_ID"),  # Optional: auto-detected from credentials
    region="us-central1",  # Default: europe-west4
)

res = client.chat.complete(
    model="mistral-small-2503",
    messages=[
        {
            "role": "user",
            "content": "Hello there!",
        }
    ],
)
print(res.choices[0].message.content)
```

--------------------------------

### JobsAPIRoutesFineTuningStartFineTuningJobRequest

Source: https://github.com/mistralai/client-python/blob/main/docs/models/jobsapiroutesfinetuningstartfinetuningjobrequest.md

Defines the request body structure for starting a fine-tuning job.

```APIDOC
## POST /v1/fine-tuning/jobs

### Description
Starts a new fine-tuning job.

### Method
POST

### Endpoint
/v1/fine-tuning/jobs

### Request Body
- **job_id** (str) - Required - The ID of the fine-tuning job.

### Request Example
{
  "job_id": "ft-j-xxxxxxxxxxxxxxxxx"
}

### Response
#### Success Response (200)
- **id** (str) - The unique identifier for the fine-tuning job.
- **object** (str) - The type of object, e.g., "fine_tuning.job".
- **created_at** (int) - Unix timestamp of job creation.
- **error** (object) - Details about any errors encountered during job creation.
- **fine_tuned_model** (str) - The name of the fine-tuned model (if applicable).
- **model** (str) - The base model used for fine-tuning.
- **object** (str) - The type of object, e.g., "fine_tuning.job".
- **result_files** (array) - List of file IDs for the results.
- **status** (str) - The current status of the job (e.g., "validating_files", "queued", "running", "succeeded", "failed", "cancelled").
- **trained_tokens** (int) - The number of tokens trained.
- **tuning_file** (str) - The ID of the training file.
- **updated_at** (int) - Unix timestamp of the last update.
- **validation_file** (str) - The ID of the validation file.

#### Response Example
{
  "id": "ft-j-xxxxxxxxxxxxxxxxx",
  "object": "fine_tuning.job",
  "created_at": 1699000000,
  "error": null,
  "fine_tuned_model": null,
  "model": "gpt-3.5-turbo",
  "result_files": [],
  "status": "validating_files",
  "trained_tokens": 0,
  "tuning_file": "file-xxxxxxxxxxxxxxxxx",
  "updated_at": 1699000000,
  "validation_file": null
}
```

--------------------------------

### Initialize ToolExecutionStartedEvent

Source: https://github.com/mistralai/client-python/blob/main/docs/models/conversationeventsdata.md

Represents the start of a tool execution.

```python
value: models.ToolExecutionStartedEvent = /* values here */
```

--------------------------------

### Get Connector by ID or Name

Source: https://github.com/mistralai/client-python/blob/main/docs/sdks/connectors/README.md

Use this snippet to retrieve a connector using its ID or name. It demonstrates how to initialize the Mistral client and make the GET request, with options to fetch associated user and customer data. Ensure the MISTRAL_API_KEY environment variable is set.

```python
from mistralai.client import Mistral
import os


with Mistral(
    api_key=os.getenv("MISTRAL_API_KEY", ""),
) as mistral:

    res = mistral.beta.connectors.get(connector_id_or_name="<value>", fetch_user_data=False, fetch_customer_data=False)

    # Handle response
    print(res)

```

--------------------------------

### Get Connector

Source: https://github.com/mistralai/client-python/blob/main/docs/sdks/connectors/README.md

Retrieves details of a specific connector.

```APIDOC
## get connector

### Description
Get a connector.

### Method
GET

### Endpoint
/connectors/{connector_id}

### Parameters
#### Path Parameters
- **connector_id** (string) - Required - The ID of the connector.
```

--------------------------------

### GET /v1/workflows/{workflow_name}/metrics

Source: https://github.com/mistralai/client-python/blob/main/docs/sdks/metrics/README.md

Retrieves comprehensive metrics for a specified workflow. You can filter metrics by providing optional start and end times in ISO 8601 format.

```APIDOC
## GET /v1/workflows/{workflow_name}/metrics

### Description
Get comprehensive metrics for a specific workflow.

### Method
GET

### Endpoint
/v1/workflows/{workflow_name}/metrics

### Parameters
#### Path Parameters
- **workflow_name** (str) - Required - The name of the workflow type to get metrics for.

#### Query Parameters
- **start_time** (date) - Optional - Filter workflows started after this time (ISO 8601).
- **end_time** (date) - Optional - Filter workflows started before this time (ISO 8601).
- **retries** (Optional[utils.RetryConfig]) - Optional - Configuration to override the default retry behavior of the client.

### Request Example
```json
{
  "example": "GET /v1/workflows/MyWorkflow/metrics"
}
```

### Response
#### Success Response (200)
- **execution_count** (int) - Total number of executions.
- **success_count** (int) - Number of successful executions.
- **error_count** (int) - Number of failed/terminated executions.
- **average_latency_ms** (float) - Average execution duration in milliseconds.
- **retry_rate** (float) - Proportion of workflows with retries.
- **latency_over_time** (dict) - Time-series data of execution durations.

#### Response Example
```json
{
  "example": "{\"execution_count\": 100, \"success_count\": 90, \"error_count\": 10, \"average_latency_ms\": 1500.5, \"retry_rate\": 0.1, \"latency_over_time\": {}}"
}
```

### Errors
- **errors.HTTPValidationError** (422) - application/json
- **errors.SDKError** (4XX, 5XX) - */*
```

--------------------------------

### Initialize SDK with Custom httpx.Client

Source: https://github.com/mistralai/client-python/blob/main/packages/gcp/README.md

Configure the SDK client with a custom httpx.Client instance to set default headers for all requests. Ensure httpx is installed.

```python
import os
from mistralai.gcp.client import MistralGCP
import httpx

http_client = httpx.Client(headers={"x-custom-header": "someValue"})
s = MistralGCP(
    project_id=os.environ.get("GCP_PROJECT_ID"),
    region="us-central1",
    client=http_client,
)
```

--------------------------------

### Append Messages to a Conversation using Python SDK

Source: https://github.com/mistralai/client-python/blob/main/docs/sdks/conversations/README.md

This example shows how to append user entries to an existing conversation and get a new completion. It demonstrates setting response format and handling the response.

```python
from mistralai.client import Mistral
import os


with Mistral(
    api_key=os.getenv("MISTRAL_API_KEY", ""),
) as mistral:

    res = mistral.beta.conversations.append(conversation_id="<id>", store=True, handoff_execution="server", completion_args={
        "response_format": {
            "type": "text",
        },
    })

    # Handle response
    print(res)

```

--------------------------------

### Initialize SDK with Custom HTTP Client Headers

Source: https://github.com/mistralai/client-python/blob/main/packages/azure/README.md

Configure a custom HTTP client with specific headers to be sent with every SDK request. This example uses `httpx.Client` for synchronous operations.

```python
from mistralai.azure.client import MistralAzure
import httpx
import os

http_client = httpx.Client(headers={"x-custom-header": "someValue"})
s = MistralAzure(
    api_key=os.environ["AZURE_API_KEY"],
    server_url=os.environ["AZURE_ENDPOINT"],
    api_version=os.environ.get("AZURE_API_VERSION", "2024-05-01-preview"),
    client=http_client,
)
```

--------------------------------

### Install additional mistralai packages

Source: https://github.com/mistralai/client-python/blob/main/README.md

Install supplementary mistralai packages, such as mistralai-workflows, which are available under the mistralai namespace.

```bash
pip install mistralai-workflows
```

--------------------------------

### Initialize SDK with API Key and Server URL

Source: https://github.com/mistralai/client-python/blob/main/packages/azure/README.md

Instantiate the Mistral Azure client using your API key and server URL. The SDK automatically handles the `api-version` query parameter. This is the standard way to authenticate and configure the client for Azure endpoints.

```python
from mistralai.azure.client import MistralAzure
import os

s = MistralAzure(
    api_key=os.environ["AZURE_API_KEY"],
    server_url=os.environ["AZURE_ENDPOINT"],
    api_version=os.environ.get("AZURE_API_VERSION", "2024-05-01-preview"),
)

res = s.chat.stream(
    messages=[
        {
            "role": "user",
            "content": "Who is the best French painter? Answer in one short sentence.",
        },
    ],
    model=os.environ["AZURE_MODEL"],
)

if res is not None:
    for event in res:
        # handle event
        print(event)
```

--------------------------------

### Create and Start a New Campaign

Source: https://github.com/mistralai/client-python/blob/main/docs/sdks/campaigns/README.md

Use this snippet to create and initiate a new campaign. Ensure you provide all required parameters like search_params, judge_id, name, description, and max_nb_events. The API key should be set via the MISTRAL_API_KEY environment variable.

```python
from mistralai.client import Mistral
import os


with Mistral(
    api_key=os.getenv("MISTRAL_API_KEY", ""),
) as mistral:

    res = mistral.beta.observability.campaigns.create(search_params={
        "filters": {
            "field": "<value>",
            "op": "lt",
            "value": "<value>",
        },
    }, judge_id="9b501b9f-3525-44a7-a51a-5352679be9ed", name="<value>", description="shakily triangular scotch requirement whether once oh", max_nb_events=232889)

    # Handle response
    print(res)


```

--------------------------------

### Get Index Schema Detail Request

Source: https://github.com/mistralai/client-python/blob/main/docs/models/getindexschemadetailv1ragindexesindexindexidschemasschemaschemaiddetailgetrequest.md

This snippet shows the structure of a request to get the index schema detail.

```APIDOC
## GET /v1/rag/indexes/{index_id}/schemas/{schema_id}/detail

### Description
Retrieves the detailed schema for a specific index and schema ID.

### Method
GET

### Endpoint
/v1/rag/indexes/{index_id}/schemas/{schema_id}/detail

### Parameters
#### Path Parameters
- **index_id** (str) - Required - The ID of the index.
- **schema_id** (str) - Required - The ID of the schema.

### Response
#### Success Response (200)
- **schema_detail** (object) - The detailed schema information.
```

--------------------------------

### Instantiate ReasoningEffort

Source: https://github.com/mistralai/client-python/blob/main/packages/gcp/docs/models/reasoningeffort.md

Demonstrates how to instantiate the ReasoningEffort model with a value. The model accepts predefined string literals.

```python
from mistralai.gcp.client.models import ReasoningEffort
value: ReasoningEffort = "none"
```

--------------------------------

### Define Response Format Examples

Source: https://github.com/mistralai/client-python/blob/main/docs/models/completionargs.md

Examples of JSON structures for defining text, JSON object, and JSON schema response formats.

```json
{
"type": "text"
}
```

```json
{
"type": "json_object"
}
```

```json
{
"type": "json_schema",
"json_schema": {
"schema": {
"properties": {
"name": {
"title": "Name",
"type": "string"
},
"authors": {
"items": {
"type": "string"
},
"title": "Authors",
"type": "array"
}
},
"required": [
"name",
"authors"
],
"title": "Book",
"type": "object",
"additionalProperties": false
},
"name": "book",
"strict": true
}
}
```

--------------------------------

### Upload a File (Synchronous)

Source: https://github.com/mistralai/client-python/blob/main/README.md

This snippet demonstrates synchronous file upload. Ensure the file 'example.file' exists and the MISTRAL_API_KEY is set.

```python
# Synchronous Example
from mistralai.client import Mistral
import os


with Mistral(
    api_key=os.getenv("MISTRAL_API_KEY", ""),
) as mistral:

    res = mistral.files.upload(file={
        "file_name": "example.file",
        "content": open("example.file", "rb"),
    }, visibility="workspace")

    # Handle response
    print(res)
```

--------------------------------

### Initialize Azure Client with Custom HTTP Client

Source: https://github.com/mistralai/client-python/blob/main/packages/azure/README-PYPI.md

Integrate a custom `httpx.Client` instance to configure timeouts, headers, and other low-level network settings for all SDK requests.

```python
from mistralai.azure.client import MistralAzure
import httpx
import os

http_client = httpx.Client(headers={"x-custom-header": "someValue"})
s = MistralAzure(
    api_key=os.environ["AZURE_API_KEY"],
    server_url=os.environ["AZURE_ENDPOINT"],
    api_version=os.environ.get("AZURE_API_VERSION", "2024-05-01-preview"),
    client=http_client,
)

```

--------------------------------

### Initialize CompletionFineTuningJobDetailsStatus

Source: https://github.com/mistralai/client-python/blob/main/docs/models/completionfinetuningjobdetailsstatus.md

Demonstrates how to import and assign a value to CompletionFineTuningJobDetailsStatus. Unrecognized values are captured as UnrecognizedStr.

```python
from mistralai.client.models import CompletionFineTuningJobDetailsStatus

# Open enum: unrecognized values are captured as UnrecognizedStr
value: CompletionFineTuningJobDetailsStatus = "QUEUED"
```

--------------------------------

### POST /v1/observability/datasets/{dataset_id}/imports/from-explorer

Source: https://github.com/mistralai/client-python/blob/main/docs/sdks/datasets/README.md

Populates a dataset with samples selected from the explorer.

```APIDOC
## POST /v1/observability/datasets/{dataset_id}/imports/from-explorer

### Description
Populates the dataset with samples from the explorer.

### Method
POST

### Endpoint
/v1/observability/datasets/{dataset_id}/imports/from-explorer

### Parameters
#### Path Parameters
- **dataset_id** (str) - Required - The ID of the dataset to import into.

#### Query Parameters
- **completion_event_ids** (list[str]) - Required - A list of completion event IDs to import from the explorer.

#### Request Body
None

### Request Example
```python
from mistralai.client import Mistral
import os


with Mistral(
    api_key=os.getenv("MISTRAL_API_KEY", ""),
) as mistral:

    res = mistral.beta.observability.datasets.import_from_explorer(dataset_id="ee1930e9-54f7-4c68-aa8a-40fe5d2a3485", completion_event_ids=[
        "<value 1>",
        "<value 2>",
        "<value 3>",
    ])

    # Handle response
    print(res)
```

### Response
#### Success Response (200)
- **DatasetImportTask** (models.DatasetImportTask) - Details of the dataset import task.

#### Response Example
```json
{
  "example": "response body"
}
```

### Errors
- **errors.ObservabilityError**: Status Code 400, 404, 408, 409, 422, Content Type application/json
- **errors.SDKError**: Status Code 4XX, 5XX, Content Type */*
```

--------------------------------

### Initialize MistralGCP Client

Source: https://github.com/mistralai/client-python/blob/main/packages/gcp/README.md

The SDK automatically constructs the Vertex AI endpoint using the provided project_id and region.

```python
import os
from mistralai.gcp.client import MistralGCP

# The SDK auto-detects credentials and builds the Vertex AI URL
s = MistralGCP(
    project_id=os.environ.get("GCP_PROJECT_ID"),  # Optional: auto-detected
    region=os.environ.get("GCP_REGION", "us-central1"),
)

res = s.chat.stream(
    messages=[
        {
            "role": "user",
            "content": "Who is the best French painter? Answer in one short sentence.",
        },
    ],
    model="mistral-small-2503",
)

if res is not None:
    for event in res:
        # handle event
        print(event)
```

--------------------------------

### Agents Completion Request Message Example

Source: https://github.com/mistralai/client-python/blob/main/docs/models/agentscompletionrequest.md

This example shows the structure of a message object within the 'messages' list for an Agents Completion Request. It includes the role and content of the prompt.

```python
{
"role": "user",
"content": "Who is the best French painter? Answer in one short sentence."
}
```

--------------------------------

### models.Tool Example

Source: https://github.com/mistralai/client-python/blob/main/packages/azure/docs/models/chatcompletionrequesttool.md

Represents a generic tool definition. Use this as a base or for custom tool implementations.

```python
value: models.Tool = /* values here */
```

--------------------------------

### Get Connector Authentication Methods

Source: https://github.com/mistralai/client-python/blob/main/docs/sdks/connectors/README.md

Retrieves the authentication methods supported by a connector.

```APIDOC
## get_authentication_methods connector

### Description
Get authentication methods for a connector.

### Method
GET

### Endpoint
/connectors/{connector_id}/auth-methods

### Parameters
#### Path Parameters
- **connector_id** (string) - Required - The ID of the connector.
```

--------------------------------

### Get Specific Conversation by ID with Python SDK

Source: https://github.com/mistralai/client-python/blob/main/docs/sdks/conversations/README.md

Retrieve a single conversation entity using its unique ID. Ensure the Mistral client is initialized and provide the `conversation_id` to the `get` method. The `conversation_id` is a required parameter.

```python
from mistralai.client import Mistral
import os


with Mistral(
    api_key=os.getenv("MISTRAL_API_KEY", ""),
) as mistral:

    res = mistral.beta.conversations.get(conversation_id="<id>")

    # Handle response
    print(res)
```

--------------------------------

### stream_workflow_execution_logs

Source: https://github.com/mistralai/client-python/blob/main/docs/sdks/executions/README.md

Stream logs for a workflow execution via SSE. This method allows for resuming streams from a specific point using `last_event_id` or starting from a timestamp with `after`. If neither is provided, it tails logs from the execution's start.

```APIDOC
## stream_workflow_execution_logs

Stream logs for a workflow execution via SSE.

If `last_event_id` is set it resumes from that cursor and takes precedence over `after`;
otherwise `after` sets a fresh stream's start point (omit both to tail from the execution start).

### Parameters

#### Path Parameters

- **execution_id** (str) - Required - The ID of the workflow execution.

#### Query Parameters

- **run_id** (Optional[str]) - Optional - Filter logs by workflow run ID.
- **activity_id** (Optional[str]) - Optional - Filter logs by activity ID.
- **after** (date) - Optional - Start a fresh stream at this timestamp (ignored when resuming via last_event_id).
- **last_event_id** (Optional[str]) - Optional - Resume from this cursor (a prior response's SSE id).

#### Request Body

This endpoint does not accept a request body.

### Response

**Union[eventstreaming.EventStream[models.StreamWorkflowExecutionLogsResponseBody], eventstreaming.EventStreamAsync[models.StreamWorkflowExecutionLogsResponseBody]]**

### Errors

- **errors.HTTPValidationError**: Status Code 422, Content Type application/json
- **errors.SDKError**: Status Code 4XX, 5XX, Content Type */*
```

--------------------------------

### Python Confirmation Model Usage

Source: https://github.com/mistralai/client-python/blob/main/docs/models/confirmation.md

Demonstrates how to use the Confirmation model with allowed values.

```python
from mistralai.client.models import Confirmation
value: Confirmation = "allow"
```

--------------------------------

### Update Connector Example

Source: https://github.com/mistralai/client-python/blob/main/docs/sdks/connectors/README.md

Use this snippet to update a connector by its ID. Ensure you have your MISTRAL_API_KEY set in your environment variables. The connector_id is required.

```python
from mistralai.client import Mistral
import os


with Mistral(
    api_key=os.getenv("MISTRAL_API_KEY", ""),
) as mistral:

    res = mistral.beta.connectors.update(connector_id="81d30634-113f-4dce-a89e-7786be2d8693")

    # Handle response
    print(res)
```

--------------------------------

### Get Connector Authentication URL

Source: https://github.com/mistralai/client-python/blob/main/docs/sdks/connectors/README.md

Retrieves the authentication URL for a specific connector.

```APIDOC
## get_auth_url connector

### Description
Get the auth URL for a connector.

### Method
GET

### Endpoint
/connectors/{connector_id}/auth-url

### Parameters
#### Path Parameters
- **connector_id** (string) - Required - The ID of the connector.
```

--------------------------------

### Initialize Mistral Client with Custom HTTP Client

Source: https://github.com/mistralai/client-python/blob/main/README-PYPI.md

Configure the SDK client with a custom `httpx.Client` instance to manage timeouts, cookies, proxies, and headers.

```python
from mistralai.client import Mistral
import httpx

http_client = httpx.Client(headers={"x-custom-header": "someValue"})
s = Mistral(client=http_client)

```

--------------------------------

### get_auth_url

Source: https://github.com/mistralai/client-python/blob/main/docs/sdks/connectors/README.md

Get the OAuth2 authorization URL for a connector to initiate user authentication.

```APIDOC
## get_auth_url

### Description
Get the OAuth2 authorization URL for a connector to initiate user authentication.

### Method
GET

### Endpoint
/v1/connectors/{connector_id_or_name}/auth_url

### Parameters
#### Path Parameters
- **connector_id_or_name** (string) - Required - The ID or name of the connector.

#### Query Parameters
- **github_installation_link** (boolean) - Optional - If true, returns a GitHub installation link.

### Request Example
```python
from mistralai.client import Mistral
import os


with Mistral(
    api_key=os.getenv("MISTRAL_API_KEY", ""),
) as mistral:

    res = mistral.beta.connectors.get_auth_url(connector_id_or_name="<value>", github_installation_link=False)

    # Handle response
    print(res)
```

### Response
#### Success Response (200)
- **auth_url** (string) - The authorization URL for the connector.

#### Response Example
```json
{
  "auth_url": "https://example.com/oauth/authorize?client_id=..."
}
```
```

--------------------------------

### Get Voice Details with Mistral AI SDK

Source: https://github.com/mistralai/client-python/blob/main/docs/sdks/voices/README.md

Use this snippet to retrieve details for a specific voice by its ID. Ensure you have your MISTRAL_API_KEY set in your environment variables. The client is initialized and then used to call the `get` method on the `voices` object.

```python
from mistralai.client import Mistral
import os


with Mistral(
    api_key=os.getenv("MISTRAL_API_KEY", ""),
) as mistral:

    res = mistral.audio.voices.get(voice_id="<id>")

    # Handle response
    print(res)

```

--------------------------------

### Start a fine-tuning job

Source: https://github.com/mistralai/client-python/blob/main/docs/sdks/finetuningjobs/README.md

Uses the fine_tuning.jobs.start method to initiate a validated fine-tuning job.

```python
from mistralai.client import Mistral
import os


with Mistral(
    api_key=os.getenv("MISTRAL_API_KEY", ""),
) as mistral:

    res = mistral.fine_tuning.jobs.start(job_id="56553e4d-0679-471e-b9ac-59a77d671103")

    # Handle response
    print(res)
```

--------------------------------

### TempoTraceAttributeArrayValue Example

Source: https://github.com/mistralai/client-python/blob/main/docs/models/tempotraceattributevalue.md

Assign an array value to a TempoTraceAttributeValue.

```python
value: models.TempoTraceAttributeArrayValue = /* values here */
```

--------------------------------

### Initialize Mistral Azure Client and Complete Chat

Source: https://github.com/mistralai/client-python/blob/main/packages/azure/docs/sdks/chat/README.md

Demonstrates how to initialize the MistralAzure client with API key, endpoint, and model, then use it to complete a chat conversation. Ensure environment variables for API key, endpoint, and model are set.

```python
from mistralai.azure.client import MistralAzure
import os

AZURE_API_KEY = os.environ["AZURE_API_KEY"]
AZURE_ENDPOINT = os.environ["AZURE_ENDPOINT"]
AZURE_MODEL = os.environ["AZURE_MODEL"]
AZURE_API_VERSION = os.environ.get("AZURE_API_VERSION", "2024-05-01-preview")

# The SDK automatically injects api-version as a query parameter
s = MistralAzure(
    api_key=AZURE_API_KEY,
    server_url=AZURE_ENDPOINT,
    api_version=AZURE_API_VERSION,
)

res = s.chat.complete(messages=[
    {
        "role": "user",
        "content": "Who is the best French painter? Answer in one short sentence.",
    },
], model=AZURE_MODEL)

if res is not None:
    # handle response
    print(res.choices[0].message.content)
```

--------------------------------

### TempoTraceAttributeBoolValue Example

Source: https://github.com/mistralai/client-python/blob/main/docs/models/tempotraceattributevalue.md

Assign a boolean value to a TempoTraceAttributeValue.

```python
value: models.TempoTraceAttributeBoolValue = /* values here */
```

--------------------------------

### Using ProcessStatus Enum

Source: https://github.com/mistralai/client-python/blob/main/docs/models/processstatus.md

Demonstrates how to import and use the ProcessStatus enum. Unrecognized string values are handled gracefully.

```python
from mistralai.client.models import ProcessStatus

# Open enum: unrecognized values are captured as UnrecognizedStr
value: ProcessStatus = "self_managed"
```

--------------------------------

### Get Connector Authentication Methods

Source: https://github.com/mistralai/client-python/blob/main/docs/sdks/connectors/README.md

Retrieve the authentication schema for a specified connector. This includes the list of supported authentication methods and their required headers. Ensure the MISTRAL_API_KEY environment variable is set.

```python
from mistralai.client import Mistral
import os


with Mistral(
    api_key=os.getenv("MISTRAL_API_KEY", ""),
) as mistral:

    res = mistral.beta.connectors.get_authentication_methods(connector_id_or_name="<value>")

    # Handle response
    print(res)

```

--------------------------------

### TempoTraceAttributeIntValue Example

Source: https://github.com/mistralai/client-python/blob/main/docs/models/tempotraceattributevalue.md

Assign an integer value to a TempoTraceAttributeValue.

```python
value: models.TempoTraceAttributeIntValue = /* values here */
```

--------------------------------

### Install Mistral AI SDK with uv

Source: https://github.com/mistralai/client-python/blob/main/README.md

Use this command to add the mistralai package to your project when using the uv package manager.

```bash
uv add mistralai
```

--------------------------------

### Configure Server URL with Mistral AI GCP Client

Source: https://github.com/mistralai/client-python/blob/main/packages/gcp/README-PYPI.md

Shows how to initialize the MistralGCP client, which automatically constructs the Vertex AI endpoint using project ID and region. This snippet also demonstrates streaming chat completions.

```python
import os
from mistralai.gcp.client import MistralGCP

# The SDK auto-detects credentials and builds the Vertex AI URL
s = MistralGCP(
    project_id=os.environ.get("GCP_PROJECT_ID"),  # Optional: auto-detected
    region=os.environ.get("GCP_REGION", "us-central1"),
)

res = s.chat.stream(
    messages=[
        {
            "role": "user",
            "content": "Who is the best French painter? Answer in one short sentence.",
        },
    ],
    model="mistral-small-2503",
)

if res is not None:
    for event in res:
        # handle event
        print(event)

```

--------------------------------

### TempoTraceAttributeStringValue Example

Source: https://github.com/mistralai/client-python/blob/main/docs/models/tempotraceattributevalue.md

Assign a string value to a TempoTraceAttributeValue.

```python
value: models.TempoTraceAttributeStringValue = /* values here */
```

--------------------------------

### Initialize SystemMessage

Source: https://github.com/mistralai/client-python/blob/main/docs/models/agentscompletionrequestmessage.md

Use this model to define system-level instructions for the agent.

```python
value: models.SystemMessage = /* values here */
```

--------------------------------

### get

Source: https://github.com/mistralai/client-python/blob/main/docs/sdks/documents/README.md

Retrieve the metadata of a specific document by its ID.

```APIDOC
## get

### Description
Retrieve the metadata of a specific document by its ID.

### Method
GET

### Endpoint
/v1/libraries/{library_id}/documents/{document_id}

### Parameters
#### Path Parameters
- **library_id** (str) - Required - The ID of the library containing the document.
- **document_id** (str) - Required - The ID of the document to retrieve.

### Request Example
```python
from mistralai.client import Mistral
import os


with Mistral(
    api_key=os.getenv("MISTRAL_API_KEY", ""),
) as mistral:

    res = mistral.beta.libraries.documents.get(library_id="5c3ca4cd-62bc-4c71-ad8a-1531ae80d078", document_id="doc_id_1")

    # Handle response
    print(res)
```

### Response
#### Success Response (200)
- **id** (str) - The ID of the document.
- **library_id** (str) - The ID of the library the document belongs to.
- **name** (str) - The name of the document.
- **status** (str) - The processing status of the document.
- **created_at** (str) - The timestamp when the document was created.
- **metadata** (dict) - The metadata associated with the document.

#### Response Example
```json
{
  "id": "doc_id_1",
  "library_id": "lib_id_1",
  "name": "my_document.pdf",
  "status": "processed",
  "created_at": "2023-01-01T10:00:00Z",
  "metadata": {"key": "value"}
}
```
```

--------------------------------

### Get Connector Authorization URL

Source: https://github.com/mistralai/client-python/blob/main/docs/sdks/connectors/README.md

Use this to obtain the OAuth2 authorization URL for a connector. This initiates the user authentication process for the connector.

```python
from mistralai.client import Mistral
import os


with Mistral(
    api_key=os.getenv("MISTRAL_API_KEY", ""),
) as mistral:

    res = mistral.beta.connectors.get_auth_url(connector_id_or_name="<value>", github_installation_link=False)

    # Handle response
    print(res)
```

--------------------------------

### Get Connector

Source: https://github.com/mistralai/client-python/blob/main/docs/sdks/connectors/README.md

Retrieves a specific connector by its unique identifier (ID or name). It allows optional fetching of associated user and customer data.

```APIDOC
## GET /v1/connectors/{connector_id_or_name}

### Description
Get a connector by its ID or name. This operation allows for optional retrieval of user and customer data associated with the connector.

### Method
GET

### Endpoint
/v1/connectors/{connector_id_or_name}

### Parameters
#### Path Parameters
- **connector_id_or_name** (string) - Required - The ID or name of the connector to retrieve.

#### Query Parameters
- **fetch_user_data** (boolean) - Optional - Fetch the user-level data associated with the connector (e.g. connection credentials).
- **fetch_customer_data** (boolean) - Optional - Fetch the customer data associated with the connector (e.g. customer secrets / config).

#### Request Body
None

### Request Example
```python
from mistralai.client import Mistral
import os


with Mistral(
    api_key=os.getenv("MISTRAL_API_KEY", ""),
) as mistral:

    res = mistral.beta.connectors.get(connector_id_or_name="<value>", fetch_user_data=False, fetch_customer_data=False)

    # Handle response
    print(res)
```

### Response
#### Success Response (200)
**models.Connector** - Represents the connector object with its associated data.

#### Response Example
```json
{
  "example": "response body"
}
```

### Errors
- **errors.HTTPValidationError** - Status Code: 422 - Content Type: application/json
- **errors.SDKError** - Status Code: 4XX, 5XX - Content Type: */*
```

--------------------------------

### GET /v1/workflows/runs

Source: https://github.com/mistralai/client-python/blob/main/docs/models/listrunsv1workflowsrunsgetresponse.md

Retrieves a list of workflow executions.

```APIDOC
## GET /v1/workflows/runs

### Description
Retrieves a list of workflow executions.

### Method
GET

### Endpoint
/v1/workflows/runs

### Response
#### Success Response (200)
- **result** (models.WorkflowExecutionListResponse) - The list of workflow executions.
```

--------------------------------

### Initialize AudioContent model

Source: https://github.com/mistralai/client-python/blob/main/docs/models/connectortoolcallresponsecontent.md

Use this model to represent audio-based content in a tool call response.

```python
value: models.AudioContent = /* values here */
```

--------------------------------

### GET /v1/conversations

Source: https://github.com/mistralai/client-python/blob/main/docs/models/agentsapiv1conversationsgetrequest.md

Retrieves metadata for a specific conversation.

```APIDOC
## GET /v1/conversations

### Description
Retrieves metadata for a specific conversation using its ID.

### Method
GET

### Endpoint
/v1/conversations

### Parameters
#### Query Parameters
- **conversation_id** (str) - Required - ID of the conversation from which we are fetching metadata.
```

--------------------------------

### Install Mistral AI SDK with Poetry

Source: https://github.com/mistralai/client-python/blob/main/README.md

Use this command to add the mistralai package to your project when using the Poetry dependency manager.

```bash
poetry add mistralai
```

--------------------------------

### CustomTaskStartedAttributesResponse

Source: https://github.com/mistralai/client-python/blob/main/docs/models/customtaskstartedattributesresponse.md

Represents the attributes of a custom task when it starts.

```APIDOC
## CustomTaskStartedAttributesResponse

Attributes for custom task started events.

### Fields

- **custom_task_id** (str) - Required - Unique identifier for the custom task within the workflow.
- **custom_task_type** (str) - Required - The type/category of the custom task (e.g., 'llm_call', 'api_request').
- **payload** (Optional[models.JSONPayloadResponse]) - Optional - A payload containing arbitrary JSON data. Used for complete state snapshots or final results. When encrypted, the value field contains base64-encoded encrypted data and encoding_options indicates the type of encryption applied.
```

--------------------------------

### Stream Agent Completions

Source: https://github.com/mistralai/client-python/blob/main/docs/sdks/agents/README.md

This example demonstrates how to stream responses from the Mistral AI agents API. It shows how to initialize the client, make a streaming request, and process the incoming events.

```APIDOC
## POST /v1/agents/completions#stream

### Description
Allows streaming responses from an agent, providing partial results as they become available. The stream is terminated with a `data: [DONE]` message. Otherwise, the server holds the request open until timeout or completion, returning the full result as JSON.

### Method
POST

### Endpoint
/v1/agents/completions#stream

### Parameters
#### Request Body
- **messages** (list[dict]) - Required - A list of message objects representing the conversation history.
- **agent_id** (str) - Required - The ID of the agent to interact with.
- **stream** (bool) - Optional - If true, enables streaming of responses. Defaults to false.
- **response_format** (dict) - Optional - Specifies the desired response format, e.g., `{"type": "text"}`.

### Request Example
```python
from mistralai.client import Mistral
import os


with Mistral(
    api_key=os.getenv("MISTRAL_API_KEY", ""),
) as mistral:

    res = mistral.agents.stream(messages=[
        {
            "role": "user",
            "content": "Who is the best French painter? Answer in one short sentence.",
        },
    ], agent_id="<id>", stream=True, response_format={
        "type": "text",
    })

    with res as event_stream:
        for event in event_stream:
            # handle event
            print(event, flush=True)
```

### Response
#### Success Response (200)
- **event** (str) - Represents a chunk of the streamed response or a control message.

#### Response Example
```
data: {"message": "..."}

data: {"message": "..."}

data: [DONE]
```
```

--------------------------------

### Initialize ResponseStartedEvent

Source: https://github.com/mistralai/client-python/blob/main/docs/models/conversationeventsdata.md

Represents the start of a response generation.

```python
value: models.ResponseStartedEvent = /* values here */
```

--------------------------------

### Initialize WandbIntegration model

Source: https://github.com/mistralai/client-python/blob/main/docs/models/createfinetuningjobrequestintegration.md

Use this structure to define a Weights & Biases integration for fine-tuning jobs.

```python
value: models.WandbIntegration = /* values here */
```

--------------------------------

### Using ShareEnum

Source: https://github.com/mistralai/client-python/blob/main/docs/models/shareenum.md

Demonstrates how to import and use the ShareEnum. Unrecognized values are handled as 'UnrecognizedStr'.

```python
from mistralai.client.models import ShareEnum

# Open enum: unrecognized values are captured as UnrecognizedStr
value: ShareEnum = "Viewer"
```

--------------------------------

### models.BuiltInConnectors Type

Source: https://github.com/mistralai/client-python/blob/main/packages/azure/docs/models/toolreferencechunktool.md

Example of assigning a value to the models.BuiltInConnectors type.

```python
value: models.BuiltInConnectors = /* values here */
```

--------------------------------

### IndexMetricsOffline Model

Source: https://github.com/mistralai/client-python/blob/main/docs/models/metricsdata.md

Example of how to declare a variable of type models.IndexMetricsOffline.

```python
value: models.IndexMetricsOffline = /* values here */
```

--------------------------------

### Initialize Mistral Client with Custom Async HTTP Client

Source: https://github.com/mistralai/client-python/blob/main/README-PYPI.md

Wrap an `httpx.AsyncClient` with a custom class implementing `AsyncHttpClient` to add logic like custom headers to all requests.

```python
from mistralai.client import Mistral
from mistralai.client.httpclient import AsyncHttpClient
import httpx

class CustomClient(AsyncHttpClient):
    client: AsyncHttpClient

    def __init__(self, client: AsyncHttpClient):
        self.client = client

    async def send(
        self,
        request: httpx.Request,
        *,
        stream: bool = False,
        auth: Union[
            httpx._types.AuthTypes, httpx._client.UseClientDefault, None
        ] = httpx.USE_CLIENT_DEFAULT,
        follow_redirects: Union[
            bool, httpx._client.UseClientDefault
        ] = httpx.USE_CLIENT_DEFAULT,
    ) -> httpx.Response:
        request.headers["Client-Level-Header"] = "added by client"

        return await self.client.send(
            request, stream=stream, auth=auth, follow_redirects=follow_redirects
        )

    def build_request(
        self,
        method: str,
        url: httpx._types.URLTypes,
        *,
        content: Optional[httpx._types.RequestContent] = None,
        data: Optional[httpx._types.RequestData] = None,
        files: Optional[httpx._types.RequestFiles] = None,
        json: Optional[Any] = None,
        params: Optional[httpx._types.QueryParamTypes] = None,
        headers: Optional[httpx._types.HeaderTypes] = None,
        cookies: Optional[httpx._types.CookieTypes] = None,
        timeout: Union[
            httpx._types.TimeoutTypes, httpx._client.UseClientDefault
        ] = httpx.USE_CLIENT_DEFAULT,
        extensions: Optional[httpx._types.RequestExtensions] = None,
    ) -> httpx.Request:
        return self.client.build_request(
            method,
            url,
            content=content,
            data=data,
            files=files,
            json=json,
            params=params,
            headers=headers,
            cookies=cookies,
            timeout=timeout,
            extensions=extensions,
        )

s = Mistral(async_client=CustomClient(httpx.AsyncClient()))

```

--------------------------------

### IndexMetricsOnline Model

Source: https://github.com/mistralai/client-python/blob/main/docs/models/metricsdata.md

Example of how to declare a variable of type models.IndexMetricsOnline.

```python
value: models.IndexMetricsOnline = /* values here */
```

--------------------------------

### POST /v1/observability/datasets/{dataset_id}/imports/from-file

Source: https://github.com/mistralai/client-python/blob/main/docs/sdks/datasets/README.md

Populate a dataset with samples from an uploaded file.

```APIDOC
## POST /v1/observability/datasets/{dataset_id}/imports/from-file

### Description
Populate the dataset with samples from an uploaded file.

### Method
POST

### Endpoint
/v1/observability/datasets/{dataset_id}/imports/from-file

### Parameters
#### Path Parameters
- **dataset_id** (str) - Required - N/A

#### Request Body
- **file_id** (str) - Required - N/A
- **retries** (Optional[utils.RetryConfig]) - Optional - Configuration to override the default retry behavior of the client.

### Request Example
```python
from mistralai.client import Mistral
import os


with Mistral(
    api_key=os.getenv("MISTRAL_API_KEY", ""),
) as mistral:

    res = mistral.beta.observability.datasets.import_from_file(dataset_id="1c96c925-cc58-4529-863d-9fe66a6f1924", file_id="<id>")

    # Handle response
    print(res)
```

### Response
#### Success Response (200)
- **models.DatasetImportTask** - Description of the task for importing the dataset.

#### Response Example
{
  "example": "response body"
}

### Errors
- **errors.ObservabilityError** - Status Code: 400, 404, 408, 409, 422, Content Type: application/json
- **errors.SDKError** - Status Code: 4XX, 5XX, Content Type: */*
```

--------------------------------

### Initialize MistralGCP Client with Custom httpx.Client

Source: https://github.com/mistralai/client-python/blob/main/packages/gcp/README-PYPI.md

Configure the MistralGCP client to use a custom httpx.Client instance with specific headers for all outgoing requests.

```python
import os
from mistralai.gcp.client import MistralGCP
import httpx

http_client = httpx.Client(headers={"x-custom-header": "someValue"})
s = MistralGCP(
    project_id=os.environ.get("GCP_PROJECT_ID"),
    region="us-central1",
    client=http_client,
)
```

--------------------------------

### ChatCompletionChoiceFinishReason Usage

Source: https://github.com/mistralai/client-python/blob/main/docs/models/chatcompletionchoicefinishreason.md

Example of how to import and use the ChatCompletionChoiceFinishReason enum.

```APIDOC
## ChatCompletionChoiceFinishReason Enum

### Description
This enum represents the reason why a chat completion sequence was stopped. It's an open enum, meaning that unrecognized values will not cause type errors and will be captured as `UnrecognizedStr`.

### Values

- `"stop"`: Indicates that the model stopped generating text because it reached a natural stopping point.
- `"length"`: Indicates that the model stopped generating text because it reached the maximum number of tokens allowed.
- `"model_length"`: Indicates that the model stopped generating text because it reached the model's maximum length.
- `"error"`: Indicates that an error occurred during generation.
- `"tool_calls"`: Indicates that the model stopped generating text to make tool calls.

### Example Usage

```python
from mistralai.client.models import ChatCompletionChoiceFinishReason

# Assigning a valid enum value
value: ChatCompletionChoiceFinishReason = "stop"

# Assigning an unrecognized value (will be captured as UnrecognizedStr)
# unrecognized_value: ChatCompletionChoiceFinishReason = "custom_reason"
```

### Request Body Example

This enum is typically used within the response of a chat completion API call, not directly in a request body. However, if you were to represent it in a request body, it would look like this:

```json
{
  "finish_reason": "stop"
}
```

### Response Example

When receiving a response from a chat completion API, the `finish_reason` field would be one of the enum values:

```json
{
  "choices": [
    {
      "message": {
        "role": "assistant",
        "content": "Hello! How can I help you today?"
      },
      "finish_reason": "stop"
    }
  ]
}
```
```

--------------------------------

### Create Agents Completions (Asynchronous)

Source: https://github.com/mistralai/client-python/blob/main/README.md

Make asynchronous agent completion requests with this example. It requires asyncio and the MISTRAL_API_KEY environment variable. Remember to set your agent_id.

```python
# Asynchronous Example
import asyncio
from mistralai.client import Mistral
import os

async def main():

    async with Mistral(
        api_key=os.getenv("MISTRAL_API_KEY", ""),
    ) as mistral:

        res = await mistral.agents.complete_async(messages=[
            {
                "role": "user",
                "content": "Who is the best French painter? Answer in one short sentence.",
            },
        ], agent_id="<id>", stream=False, response_format={
            "type": "text",
        })

        # Handle response
        print(res)

asyncio.run(main())
```

--------------------------------

### Beta Agents Get

Source: https://github.com/mistralai/client-python/blob/main/README.md

Retrieves a specific agent entity.

```APIDOC
## Beta.Agents.get

### Description
Retrieve an agent entity retrieves the details of a specific agent.

### Method
Not specified (assumed to be a client SDK method call)

### Endpoint
Not specified

### Parameters
Not specified

### Request Example
Not specified

### Response
Not specified
```

--------------------------------

### List Agents - Python

Source: https://github.com/mistralai/client-python/blob/main/docs/sdks/betaagents/README.md

Use this snippet to list available agents. It initializes the Mistral client with an API key and calls the agents.list method with pagination parameters.

```python
from mistralai.client import Mistral
import os


with Mistral(
    api_key=os.getenv("MISTRAL_API_KEY", ""),
) as mistral:

    res = mistral.beta.agents.list(page=0, page_size=20)

    # Handle response
    print(res)
```

--------------------------------

### Audio Voices Get

Source: https://github.com/mistralai/client-python/blob/main/README.md

Retrieves details for a specific voice.

```APIDOC
## Audio.Voices.get

### Description
Get voice details for a specific voice.

### Method
Not specified (assumed to be a client SDK method call)

### Endpoint
Not specified

### Parameters
Not specified

### Request Example
Not specified

### Response
Not specified
```

--------------------------------

### Using ClassifierFineTuningJobDetailsStatus

Source: https://github.com/mistralai/client-python/blob/main/docs/models/classifierfinetuningjobdetailsstatus.md

Demonstrates how to import and use ClassifierFineTuningJobDetailsStatus. Unrecognized values are captured as UnrecognizedStr.

```python
from mistralai.client.models import ClassifierFineTuningJobDetailsStatus

# Open enum: unrecognized values are captured as UnrecognizedStr
value: ClassifierFineTuningJobDetailsStatus = "QUEUED"
```