langflow-ai/langflow · error · HTTPException

Could not load flow module: {flow_path}

Error message

Could not load flow module: {flow_path}

What it means

Raised when importlib.util.spec_from_file_location returns None (or a spec without a loader) for a .py flow file. This means Python's import machinery could not create a loader for the path at all — classically because the file extension is not importable or the path is not a regular file. HTTP 500 with the offending path echoed.

Source

Thrown at src/backend/base/langflow/agentic/services/helpers/flow_loader.py:137

    Args:
        flow_path: Path to the Python flow file.
        provider: Optional model provider (e.g., "OpenAI").
        model_name: Optional model name (e.g., "gpt-4o-mini").
        api_key_var: Optional API key variable name.
        provider_vars: Resolved request variables; ``ITERATIONS_LIMIT`` is forwarded
            to ``get_graph(iterations_limit=...)`` when the flow accepts it.

    Returns:
        Graph: The loaded and configured graph.

    Raises:
        HTTPException: If the flow file cannot be loaded or executed.
    """
    module_name = flow_path.stem
    spec = importlib.util.spec_from_file_location(module_name, flow_path)
    if spec is None or spec.loader is None:
        raise HTTPException(status_code=500, detail=f"Could not load flow module: {flow_path}")

    module = importlib.util.module_from_spec(spec)
    sys.modules[module_name] = module

    try:
        with _temporary_sys_path(str(flow_path.parent)):
            spec.loader.exec_module(module)
    except Exception as e:
        if module_name in sys.modules:
            del sys.modules[module_name]
        logger.error(f"Error loading Python flow module: {e}")
        raise HTTPException(status_code=500, detail=f"Error loading flow module: {e}") from e

    if not hasattr(module, "get_graph"):
        # Fallback: check for 'graph' variable for backward compatibility
        if hasattr(module, "graph"):
            graph = module.graph
            validate_flow_for_current_settings(graph)

View on GitHub (pinned to 976ec789d2)

Solutions

1. Verify FLOWS_BASE_PATH/<name> is a regular file (not a directory) and still exists at call time.
2. Redeploy the flow file if it was removed; avoid deleting files while requests are in flight (write-then-rename deployments).
3. If it recurs, inspect the path in the error detail against the actual filesystem layout.

import os, pathlib

def importable_py_flow(path: str) -> bool:
    p = pathlib.Path(path)
    return p.suffix == '.py' and p.is_file() and not p.is_dir() and os.access(p, os.R_OK)

On 500 'Could not load flow module', verify the path from the detail is a regular file, redeploy if it vanished, and retry once after confirmation.

When it happens

Trigger: A flow resolved to a .py path that reached _load_graph_from_python but whose path cannot yield an import spec — e.g. a file with a .py name that is actually a directory, a dangling state after the file was deleted between resolve and import, or an exotic filesystem where FileLoader is unavailable.

Common situations: Race with a deploy job deleting/renaming files mid-request; a directory accidentally named 'flow.py' inside the flows folder; unusual mounts (some network filesystems) confusing the loader.

Related errors

• Error loading flow module: {e}
• An error occurred while executing the flow.
• An internal error occurred while executing the flow.
• An error occurred while preparing the flow.
• Invalid flow filename

AI-assisted analysis of langflow-ai/langflow@976ec789d2 (2026-08-14). Data as JSON: /api/errors/b1866f44521fdf7e. Report an issue: GitHub.