langchain-ai/langgraph · error · NotImplementedError

Unsupported callback types: {type(base)}, {type(new)}

Error message

Unsupported callback types: {type(base)}, {type(new)}

What it means

Error "Unsupported callback types: {type(base)}, {type(new)}" thrown in langchain-ai/langgraph.

Source

Thrown at libs/langgraph/langgraph/_internal/_config.py:144

    if base is None:
        return new.copy() if isinstance(new, (list, BaseCallbackManager)) else new
    if isinstance(new, list):
        if isinstance(base, list):
            return base + new
        if isinstance(base, BaseCallbackManager):
            mngr = base.copy()
            for cb in new:
                mngr.add_handler(cb, inherit=True)
            return mngr
    elif isinstance(new, BaseCallbackManager):
        if isinstance(base, list):
            mngr = new.copy()
            for cb in base:
                mngr.add_handler(cb, inherit=True)
            return mngr
        if isinstance(base, BaseCallbackManager):
            return base.merge(new)
    raise NotImplementedError(f"Unsupported callback types: {type(base)}, {type(new)}")


def merge_configs(*configs: RunnableConfig | None) -> RunnableConfig:
    """Merge multiple configs into one.

    Args:
        *configs: The configs to merge.

    Returns:
        RunnableConfig: The merged config.
    """
    base: RunnableConfig = {}
    # Even though the keys aren't literals, this is correct
    # because both dicts are the same type
    for config in configs:
        if config is None:
            continue
        for key, value in config.items():

View on GitHub (pinned to 38031739e5)

When it happens

Trigger: Thrown at libs/langgraph/langgraph/_internal/_config.py:144 when the library encounters an invalid state.

Common situations: See trigger scenarios.

AI-assisted analysis of langchain-ai/langgraph@38031739e5 (2026-08-26). Data as JSON: /api/errors/399291b1aba5c8c9. Report an issue: GitHub.