BerriAI/litellm · error · ValueError

Invalid MIME type: {mime_type}

Error message

Invalid MIME type: {mime_type}

What it means

Raised when the resolved MIME type (from the filename, file object name, or document['mime_type'] override) is not a valid MIME string, so it cannot be embedded in the base64 data URI for the OCR request.

Source

Thrown at litellm/ocr/main.py:549

            file_name = getattr(file_input, "name", None)
            if file_name:
                mime_type = get_mime_type(file_name)
        file_bytes = file_input.read()
        if isinstance(file_bytes, str):
            file_bytes = file_bytes.encode("utf-8")
    else:
        raise ValueError(
            f"Unsupported file input type: {type(file_input)}. Expected pathlib.Path, bytes, or a file-like object."
        )

    if not file_bytes:
        raise ValueError("File is empty or could not be read")

    if "mime_type" in document:
        mime_type = document["mime_type"]

    if not _MIME_PATTERN.match(mime_type):
        raise ValueError(f"Invalid MIME type: {mime_type}")

    base64_data: Final = base64.b64encode(file_bytes).decode("utf-8")
    data_uri: Final = f"data:{mime_type};base64,{base64_data}"

    if mime_type.startswith("image/"):
        verbose_logger.debug(
            "OCR file input: Converted file to image_url data URI (mime=%s, size=%s bytes, name=%s)",
            mime_type,
            len(file_bytes),
            file_name,
        )
        return {"type": "image_url", "image_url": data_uri}

    verbose_logger.debug(
        "OCR file input: Converted file to document_url data URI (mime=%s, size=%s bytes, name=%s)",
        mime_type,
        len(file_bytes),
        file_name,

View on GitHub (pinned to 77b7c6c40c)

Solutions

1. Provide a supported MIME type (e.g. application/pdf, image/png).
2. Check the file extension matches its actual content.

Example fix

mime_type='application/pdf'

When it happens

Trigger: Thrown at litellm/ocr/main.py:549 when the library encounters an invalid state.

Common situations: See trigger scenarios.

AI-assisted analysis of BerriAI/litellm@77b7c6c40c (2026-08-18). Data as JSON: /api/errors/1cba2a836fe135b0. Report an issue: GitHub.