Skip to content

AzureOpenAILLM

AzureOpenAILLM

Bases: OpenAILLM

Azure OpenAI LLM implementation running the async API client.

Attributes:

Name Type Description
model

the model name to use for the LLM i.e. the name of the Azure deployment.
base_url Optional[RuntimeParameter[str]]

the base URL to use for the Azure OpenAI API can be set with AZURE_OPENAI_ENDPOINT. Defaults to None which means that the value set for the environment variable AZURE_OPENAI_ENDPOINT will be used, or None if not set.
api_key Optional[RuntimeParameter[SecretStr]]

the API key to authenticate the requests to the Azure OpenAI API. Defaults to None which means that the value set for the environment variable AZURE_OPENAI_API_KEY will be used, or None if not set.
api_version Optional[RuntimeParameter[str]]

the API version to use for the Azure OpenAI API. Defaults to None which means that the value set for the environment variable OPENAI_API_VERSION will be used, or None if not set.
Icon

:simple-microsoftazure:

Examples:

Generate text:

```python
from distilabel.llms import AzureOpenAILLM

llm = AzureOpenAILLM(model="gpt-4-turbo", api_key="api.key")

llm.load()

# Synchrounous request
output = llm.generate(inputs=[[{"role": "user", "content": "Hello world!"}]])

# Asynchronous request
output = await llm.agenerate(input=[{"role": "user", "content": "Hello world!"}])
```

Generate text from a custom endpoint following the OpenAI API:

```python
from distilabel.llms import AzureOpenAILLM

llm = AzureOpenAILLM(
    model="prometheus-eval/prometheus-7b-v2.0",
    base_url=r"http://localhost:8080/v1"
)

llm.load()

# Synchronous request
output = llm.generate(inputs=[[{"role": "user", "content": "Hello world!"}]])

# Asynchronous request
output = await llm.agenerate(input=[{"role": "user", "content": "Hello world!"}])
```

Generate structured data:

```python
from pydantic import BaseModel
from distilabel.llms import AzureOpenAILLM

class User(BaseModel):
    name: str
    last_name: str
    id: int

llm = AzureOpenAILLM(
    model="gpt-4-turbo",
    api_key="api.key",
    structured_output={"schema": User}
)

llm.load()

output = llm.generate(inputs=[[{"role": "user", "content": "Create a user profile for the following marathon"}]])
```
Source code in src/distilabel/llms/azure.py 
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
class AzureOpenAILLM(OpenAILLM):
    """Azure OpenAI LLM implementation running the async API client.

    Attributes:
        model: the model name to use for the LLM i.e. the name of the Azure deployment.
        base_url: the base URL to use for the Azure OpenAI API can be set with `AZURE_OPENAI_ENDPOINT`.
            Defaults to `None` which means that the value set for the environment variable
            `AZURE_OPENAI_ENDPOINT` will be used, or `None` if not set.
        api_key: the API key to authenticate the requests to the Azure OpenAI API. Defaults to `None`
            which means that the value set for the environment variable `AZURE_OPENAI_API_KEY` will be
            used, or `None` if not set.
        api_version: the API version to use for the Azure OpenAI API. Defaults to `None` which means
            that the value set for the environment variable `OPENAI_API_VERSION` will be used, or
            `None` if not set.

    Icon:
        `:simple-microsoftazure:`

    Examples:

        Generate text:

        ```python
        from distilabel.llms import AzureOpenAILLM

        llm = AzureOpenAILLM(model="gpt-4-turbo", api_key="api.key")

        llm.load()

        # Synchrounous request
        output = llm.generate(inputs=[[{"role": "user", "content": "Hello world!"}]])

        # Asynchronous request
        output = await llm.agenerate(input=[{"role": "user", "content": "Hello world!"}])
        ```

        Generate text from a custom endpoint following the OpenAI API:

        ```python
        from distilabel.llms import AzureOpenAILLM

        llm = AzureOpenAILLM(
            model="prometheus-eval/prometheus-7b-v2.0",
            base_url=r"http://localhost:8080/v1"
        )

        llm.load()

        # Synchronous request
        output = llm.generate(inputs=[[{"role": "user", "content": "Hello world!"}]])

        # Asynchronous request
        output = await llm.agenerate(input=[{"role": "user", "content": "Hello world!"}])
        ```

        Generate structured data:

        ```python
        from pydantic import BaseModel
        from distilabel.llms import AzureOpenAILLM

        class User(BaseModel):
            name: str
            last_name: str
            id: int

        llm = AzureOpenAILLM(
            model="gpt-4-turbo",
            api_key="api.key",
            structured_output={"schema": User}
        )

        llm.load()

        output = llm.generate(inputs=[[{"role": "user", "content": "Create a user profile for the following marathon"}]])
        ```
    """

    base_url: Optional[RuntimeParameter[str]] = Field(
        default_factory=lambda: os.getenv(_AZURE_OPENAI_ENDPOINT_ENV_VAR_NAME),
        description="The base URL to use for the Azure OpenAI API requests i.e. the Azure OpenAI endpoint.",
    )
    api_key: Optional[RuntimeParameter[SecretStr]] = Field(
        default_factory=lambda: os.getenv(_AZURE_OPENAI_API_KEY_ENV_VAR_NAME),
        description="The API key to authenticate the requests to the Azure OpenAI API.",
    )

    api_version: Optional[RuntimeParameter[str]] = Field(
        default_factory=lambda: os.getenv("OPENAI_API_VERSION"),
        description="The API version to use for the Azure OpenAI API.",
    )

    _base_url_env_var: str = PrivateAttr(_AZURE_OPENAI_ENDPOINT_ENV_VAR_NAME)
    _api_key_env_var: str = PrivateAttr(_AZURE_OPENAI_API_KEY_ENV_VAR_NAME)
    _aclient: Optional["AsyncAzureOpenAI"] = PrivateAttr(...)  # type: ignore

    @override
    def load(self) -> None:
        """Loads the `AsyncAzureOpenAI` client to benefit from async requests."""
        # This is a workaround to avoid the `OpenAILLM` calling the _prepare_structured_output
        # in the load method before we have the proper client.
        with patch(
            "distilabel.llms.openai.OpenAILLM._prepare_structured_output", lambda x: x
        ):
            super().load()

        try:
            from openai import AsyncAzureOpenAI
        except ImportError as ie:
            raise ImportError(
                "OpenAI Python client is not installed. Please install it using"
                " `pip install openai`."
            ) from ie

        if self.api_key is None:
            raise ValueError(
                f"To use `{self.__class__.__name__}` an API key must be provided via `api_key`"
                f" attribute or runtime parameter, or set the environment variable `{self._api_key_env_var}`."
            )

        # TODO: May be worth adding the AD auth too? Also the `organization`?
        self._aclient = AsyncAzureOpenAI(  # type: ignore
            azure_endpoint=self.base_url,  # type: ignore
            azure_deployment=self.model,
            api_version=self.api_version,
            api_key=self.api_key.get_secret_value(),
            max_retries=self.max_retries,  # type: ignore
            timeout=self.timeout,
        )

        if self.structured_output:
            self._prepare_structured_output(self.structured_output)

load()

Loads the AsyncAzureOpenAI client to benefit from async requests.

Source code in src/distilabel/llms/azure.py 
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
@override
def load(self) -> None:
    """Loads the `AsyncAzureOpenAI` client to benefit from async requests."""
    # This is a workaround to avoid the `OpenAILLM` calling the _prepare_structured_output
    # in the load method before we have the proper client.
    with patch(
        "distilabel.llms.openai.OpenAILLM._prepare_structured_output", lambda x: x
    ):
        super().load()

    try:
        from openai import AsyncAzureOpenAI
    except ImportError as ie:
        raise ImportError(
            "OpenAI Python client is not installed. Please install it using"
            " `pip install openai`."
        ) from ie

    if self.api_key is None:
        raise ValueError(
            f"To use `{self.__class__.__name__}` an API key must be provided via `api_key`"
            f" attribute or runtime parameter, or set the environment variable `{self._api_key_env_var}`."
        )

    # TODO: May be worth adding the AD auth too? Also the `organization`?
    self._aclient = AsyncAzureOpenAI(  # type: ignore
        azure_endpoint=self.base_url,  # type: ignore
        azure_deployment=self.model,
        api_version=self.api_version,
        api_key=self.api_key.get_secret_value(),
        max_retries=self.max_retries,  # type: ignore
        timeout=self.timeout,
    )

    if self.structured_output:
        self._prepare_structured_output(self.structured_output)