BerriAI/litellm · error · ValueError

Unsupported storage backend type: {backend_type}. Supported

Error message

Unsupported storage backend type: {backend_type}. Supported types: azure_storage

What it means

Raised by the storage backend factory when the requested backend_type string is anything other than 'azure_storage'. The factory is intentionally minimal: it maps 'azure_storage' -> AzureBlobStorageBackend() and rejects everything else, so any typo or unsupported backend name (s3, gcs, local) fails immediately.

Source

Thrown at litellm/llms/base_llm/files/storage_backend_factory.py:37

    Backends are configured using the same environment variables as their
    corresponding callbacks. For example, "azure_storage" uses the same
    env vars as AzureBlobStorageLogger.

    Args:
        backend_type: Backend type identifier (e.g., "azure_storage")

    Returns:
        BaseFileStorageBackend: Instance of the appropriate storage backend

    Raises:
        ValueError: If backend_type is not supported
    """
    verbose_logger.debug("Creating storage backend: type=%s", backend_type)

    if backend_type == "azure_storage":
        return AzureBlobStorageBackend()
    else:
        raise ValueError(f"Unsupported storage backend type: {backend_type}. Supported types: azure_storage")

View on GitHub (pinned to 6c2dcb801b)

Solutions

1. Set backend_type exactly to 'azure_storage'.
2. Check your litellm version (pip show litellm) — if you need another backend, upgrade to a version whose factory supports it.
3. For non-Azure storage, upload files via a pre-signed URL flow or your own storage layer instead of this factory.
4. If adding a custom backend, register it in the factory (add an elif branch returning your BaseFileStorageBackend subclass).

Example fix

# before
backend = create_storage_backend(backend_type='s3')  # ValueError

# after
backend = create_storage_backend(backend_type='azure_storage')

SUPPORTED_BACKENDS = {'azure_storage'}

if backend_type not in SUPPORTED_BACKENDS:
    raise ConfigError(f'{backend_type!r} unsupported; choose from {SUPPORTED_BACKENDS}')

def is_supported_backend(value: str) -> bool:
    """Type guard for storage backend factory inputs."""
    return value in {'azure_storage'}

When it happens

Trigger: Passing backend_type='s3', 'S3', 'azure-blob', 'Azure_Storage' or any other string to the factory (via proxy file-management config or direct call); version drift where config written for a newer litellm mentions backends this version does not ship.

Common situations: Config copied from docs of a different litellm version; assuming S3/GCS support exists because constants exist elsewhere; case-sensitive string mismatches between config value and factory check.

Related errors

• Error: {response.status_code} - {response.text}
• Missing Authorization header
• Invalid bearer token
• Invalid API key
• Prompt '{prompt_id}' not found

AI-assisted analysis of BerriAI/litellm@6c2dcb801b (2026-08-15). Data as JSON: /api/errors/eb43df68f44a580c. Report an issue: GitHub.