BerriAI/litellm · error · ValueError

AzureOpenAI client is not initialized. Make sure api_key is

Error message

AzureOpenAI client is not initialized. Make sure api_key is passed or OPENAI_API_KEY is set in the environment.

What it means

In the Azure files handler, get_azure_openai_client() is asked to build a files-capable client from api_key/api_base/api_version/litellm_params. If it returns None — which happens when no api_key is available and the OPENAI_API_KEY env var is unset — the handler raises ValueError, since Azure file operations need an authenticated client. The isinstance checks that follow distinguish sync/async mismatches.

Source

Thrown at litellm/llms/azure/files/handler.py:74

        create_file_data: CreateFileRequest,
        api_base: str | None,
        api_key: str | None,
        api_version: str | None,
        timeout: float | httpx.Timeout,
        max_retries: int | None,
        client: AzureOpenAI | AsyncAzureOpenAI | OpenAI | AsyncOpenAI | None = None,
        litellm_params: dict | None = None,
    ) -> OpenAIFileObject | Coroutine[Any, Any, OpenAIFileObject]:
        openai_client: AzureOpenAI | AsyncAzureOpenAI | OpenAI | AsyncOpenAI | None = self.get_azure_openai_client(
            litellm_params=litellm_params or {},
            api_key=api_key,
            api_base=api_base,
            api_version=api_version,
            client=client,
            _is_async=_is_async,
        )
        if openai_client is None:
            raise ValueError(
                "AzureOpenAI client is not initialized. Make sure api_key is passed or OPENAI_API_KEY is set in the environment."
            )

        if _is_async is True:
            if not isinstance(openai_client, (AsyncAzureOpenAI, AsyncOpenAI)):
                raise ValueError(
                    "AzureOpenAI client is not an instance of AsyncAzureOpenAI. Make sure you passed an AsyncAzureOpenAI client."
                )
            return self.acreate_file(create_file_data=create_file_data, openai_client=openai_client)
        response: Final = cast(AzureOpenAI | OpenAI, openai_client).files.create(
            **self._prepare_create_file_data(create_file_data)
        )
        return OpenAIFileObject(**response.model_dump())

    async def afile_content(
        self,
        file_content_request: FileContentRequest,
        openai_client: AsyncAzureOpenAI | AsyncOpenAI,

View on GitHub (pinned to 6c2dcb801b)

Solutions

1. Pass api_key explicitly to the files call, or export OPENAI_API_KEY/AZURE_API_KEY in the environment.
2. Verify with: python -c "import os; print(os.getenv('OPENAI_API_KEY'))" before running the upload.
3. In the proxy, ensure the files-capable model entry includes api_key (litellm_params).
4. If using azure_ad_token-based auth, confirm the token provider path is configured so the client factory can authenticate.

Example fix

# before
file_obj = azure_files.create_file(file=open("f.jsonl","rb"), purpose="fine-tune")  # no key anywhere

# after
file_obj = azure_files.create_file(
    file=open("f.jsonl","rb"),
    purpose="fine-tune",
    api_key=os.environ["AZURE_API_KEY"],
    api_base="https://my-resource.openai.azure.com",
    api_version="2024-10-21",
)

import os

def validate_files_config(api_key: str | None) -> None:
    if not (api_key or os.getenv("OPENAI_API_KEY") or os.getenv("AZURE_API_KEY")):
        raise ConfigError("No API key for Azure files call; pass api_key or set OPENAI_API_KEY")

from openai import AzureOpenAI, AsyncAzureOpenAI

def is_files_client(c: object, is_async: bool) -> bool:
    return isinstance(c, AsyncAzureOpenAI) if is_async else isinstance(c, AzureOpenAI)

try:
    f = azure_files.create_file(file=fh, purpose="fine-tune", api_key=key, api_base=base, api_version=ver)
except ValueError as e:
    if "not initialized" in str(e):
        raise ConfigError("Missing API key for Azure files client") from e
    raise

When it happens

Trigger: Calling AzureOpenAI_Files.create_file() without api_key and without OPENAI_API_KEY (or AZURE_API_KEY as resolved by the client factory) in the environment; passing api_base/api_version only, with the key lookup failing in the secret manager.

Common situations: Batch/fine-tuning file-upload scripts that set AZURE_API_BASE and AZURE_API_VERSION but forget the key; proxy setups where files endpoints are configured with only a base URL; CI environments missing the secret env var.

Related errors

• api_key is required for Azure AI Speech transcription.
• AZURE_CLIENT_ID and AZURE_TENANT_ID must be set
• GEMINI_API_KEY is required for Google AI Studio file operati
• api_key is required
• Error: {response.status_code} - {response.text}

AI-assisted analysis of BerriAI/litellm@6c2dcb801b (2026-08-15). Data as JSON: /api/errors/9ec953ac50464242. Report an issue: GitHub.