BerriAI/litellm · error · ValueError

Unified id does not contain {marker!r}: {file_id[:80]!r}

Error message

Unified id does not contain {marker!r}: {file_id[:80]!r}

What it means

get_output_file_id_from_unified_file_id parses the internal unified file id format, which embeds segments like 'llm_output_file_id,<id>;'. If that marker is absent — corrupt, truncated, or a foreign/unmanaged id that happened to decode — a ValueError is raised showing the first 80 chars for diagnosis. The id was already accepted as base64 unified id, so this signals format corruption rather than a wrong id type.

Source

Thrown at enterprise/litellm_enterprise/proxy/hooks/managed_files.py:1139

    def get_unified_output_file_id(self, output_file_id: str, model_id: str, model_name: Optional[str]) -> str:
        deterministic_uuid: Final = uuid5(uuid5(NAMESPACE_URL, model_id), output_file_id)
        unified_output_file_id = SpecialEnums.LITELLM_MANAGED_FILE_COMPLETE_STR.value.format(
            "application/json",
            str(deterministic_uuid),
            model_name or "",
            output_file_id,
            model_id,
        )
        return base64.urlsafe_b64encode(unified_output_file_id.encode()).decode().rstrip("=")

    def get_model_id_from_unified_file_id(self, file_id: str) -> str:
        return file_id.split("llm_output_file_model_id,")[1].split(";")[0]

    def get_output_file_id_from_unified_file_id(self, file_id: str) -> str:
        marker = "llm_output_file_id,"
        if marker not in file_id:
            raise ValueError(f"Unified id does not contain {marker!r}: {file_id[:80]!r}")
        return file_id.split(marker, 1)[1].split(";")[0]

    async def async_post_call_success_hook(
        self, data: Dict, user_api_key_dict: UserAPIKeyAuth, response: LLMResponseTypes
    ) -> LLMResponseTypes:
        if isinstance(response, LiteLLMBatch):
            ## Check if unified_file_id is in the response
            unified_file_id = response._hidden_params.get("unified_file_id")  # managed file id
            unified_batch_id = response._hidden_params.get("unified_batch_id")  # managed batch id
            model_id = cast(Optional[str], response._hidden_params.get("model_id"))
            model_name = cast(Optional[str], response._hidden_params.get("model_name"))

            resolved_model_name = resolve_managed_output_file_model_name(
                unified_input_file_id=unified_file_id if isinstance(unified_file_id, str) else response.input_file_id,
                fallback_model_name=model_name,
            )
            original_response_id = response.id

View on GitHub (pinned to 6c2dcb801b)

Solutions

1. Use the output file id exactly as returned in the batch completion payload, not reconstructed or stored copies
2. Verify stored id columns/properties can hold the full base64 string without truncation
3. Confirm you are passing an output-file unified id (contains llm_output_file_id) and not a batch or input-file id
4. Recreate the batch if its ids predate the current unified-id format

Example fix

# before
file_id = batch_response.id  # batch id, wrong parser path

# after
file_id = batch_response.output_file_id  # correct output-file unified id

OUTPUT_MARKER = "llm_output_file_id,"

def is_output_file_unified_id(fid: str) -> bool:
    import base64
    try:
        raw = base64.urlsafe_b64decode(fid + "=" * (-len(fid) % 4)).decode()
        return OUTPUT_MARKER in raw
    except Exception:
        return False

assert is_output_file_unified_id(fid), "not an output-file unified id"

def is_output_file_unified_id(file_id: str) -> bool:
    import base64
    try:
        raw = base64.urlsafe_b64decode(file_id + "=" * (-len(file_id) % 4)).decode()
        return "llm_output_file_id," in raw
    except Exception:
        return False

try:
    get_output_file_id_from_unified_file_id(fid)
except ValueError as e:
    # id lacks the output-file segment; refetch the correct id from the batch object
    raise TypeError(f"{fid} is not an output-file unified id") from e

When it happens

Trigger: A base64 string that decodes but lacks the output-file segment (batch output file ids from an older format or hand-assembled); ids truncated by storage columns or URL handling; decoding a unified batch id with the file-id parser.

Common situations: Schema migrations that shortened id columns; middleware trimming long ids; mixing id types (batch id vs output file id) in client code after API changes.

Related errors

• LiteLLM Managed {accessor_key} with id={retrieve_object_id}
• LiteLLM Managed File object with id={file_id} has no file_ob
• LiteLLM Managed File object with id={file_id} not found. Che
• file_id is required in file_content_request
• contents of file are None

AI-assisted analysis of BerriAI/litellm@6c2dcb801b (2026-08-15). Data as JSON: /api/errors/653f662f28726122. Report an issue: GitHub.