BerriAI/litellm · error · ValueError

api_base is required

Error message

api_base is required

What it means

GeminiFilesConfig.get_complete_url() builds the resumable-upload URL {api_base}/upload/v1beta/files. It resolves api_base via self.get_api_base() (explicit argument, then GEMINI_API_BASE env); if the result is falsy it raises 'api_base is required' before constructing the URL.

Source

Thrown at litellm/llms/gemini/files/transformation.py:82

        self,
        api_base: str | None,
        api_key: str | None,
        model: str,
        optional_params: dict,
        litellm_params: dict,
        stream: bool | None = None,
    ) -> str:
        """
        OPTIONAL

        Get the complete url for the request

        Some providers need `model` in `api_base`
        """
        endpoint: Final = "upload/v1beta/files"
        api_base = self.get_api_base(api_base)
        if not api_base:
            raise ValueError("api_base is required")

        # Get API key from multiple sources
        final_api_key: Final = api_key or litellm_params.get("api_key") or self.get_api_key()
        if not final_api_key:
            raise ValueError("api_key is required")

        url: Final = f"{api_base}/{endpoint}"
        return url

    def get_supported_openai_params(self, model: str) -> list[OpenAICreateFileRequestOptionalParams]:
        return []

    def map_openai_params(
        self,
        non_default_params: dict,
        optional_params: dict,
        model: str,
        drop_params: bool,

View on GitHub (pinned to 6c2dcb801b)

Solutions

1. Export GEMINI_API_BASE=https://generativelanguage.googleapis.com (the standard AI Studio base).
2. Or pass api_base explicitly when invoking the file operation.
3. Audit code that forwards api_base for accidental empty-string values, which also fail the check.

Example fix

# before
os.environ.pop("GEMINI_API_BASE", None)
litellm.create_file(model="gemini/", file=f)  # api_base unresolvable

# after
os.environ["GEMINI_API_BASE"] = "https://generativelanguage.googleapis.com"
litellm.create_file(model="gemini/", file=f)

import os

def gemini_files_base() -> str:
    base = os.getenv("GEMINI_API_BASE") or "https://generativelanguage.googleapis.com"
    if not base:
        raise RuntimeError("GEMINI_API_BASE must be a non-empty URL")
    return base

def is_valid_api_base(base: object) -> bool:
    return isinstance(base, str) and base.strip() != ""

try:
    litellm.create_file(model="gemini/", file=f, purpose="user_data")
except ValueError as e:
    if "api_base is required" in str(e):
        os.environ["GEMINI_API_BASE"] = "https://generativelanguage.googleapis.com"
        litellm.create_file(model="gemini/", file=f, purpose="user_data")
    else:
        raise

When it happens

Trigger: File operations invoked with api_base=None/'' and no GEMINI_API_BASE environment variable set (e.g. an environment where the default base was cleared or a custom integration forgot to pass it).

Common situations: Custom orchestration code that forwards api_base=None explicitly; GEMINI_API_BASE set to an empty string in .env; a test harness that strips env vars and calls the transformation directly.

Related errors

• api_base is required
• GEMINI_API_KEY is required for Google AI Studio file operati
• api_key is required
• LLM Router not initialized. Ensure models added to proxy.
• API base is required for OpenAI image variations

AI-assisted analysis of BerriAI/litellm@6c2dcb801b (2026-08-15). Data as JSON: /api/errors/6ec175094a1b2c12. Report an issue: GitHub.