langchain-ai/langgraph · error · InvalidUpdateError

Cannot apply multiple updates when updating as input

Error message

Cannot apply multiple updates when updating as input

What it means

Raised when update_state is called with as_node semantics that require applying updates as graph input while more than one update item was supplied. Multiple updates can only be attributed to specific nodes; input-style application accepts exactly one. Reduce to a single update or specify a node.

Solutions

1. Apply updates one at a time when using as_node equal to the input
2. Provide explicit as_node values for each update

When it happens

Trigger: Thrown at libs/langgraph/langgraph/pregel/main.py:1749 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/ce5d92293f44fe76. Report an issue: GitHub.

Appendix: source

Thrown at libs/langgraph/langgraph/pregel/main.py:1749

                    create_checkpoint(checkpoint, channels, step),
                    {
                        "source": "update",
                        "step": step + 1,
                        "parents": saved.metadata.get("parents", {}) if saved else {},
                    },
                    get_new_channel_versions(
                        checkpoint_previous_versions,
                        checkpoint["channel_versions"],
                    ),
                )
                return patch_checkpoint_map(
                    next_config, saved.metadata if saved else None
                )

            # act as an input
            if as_node == INPUT:
                if len(updates) > 1:
                    raise InvalidUpdateError(
                        "Cannot apply multiple updates when updating as input"
                    )

                if input_writes := deque(map_input(self.input_channels, values)):
                    apply_writes(
                        checkpoint,
                        channels,
                        [PregelTaskWrites((), INPUT, input_writes, [])],
                        checkpointer.get_next_version,
                        self.trigger_to_nodes,
                    )

                    # apply input write to channels
                    next_step = (
                        step + 1
                        if saved and saved.metadata.get("step") is not None
                        else -1
                    )

View on GitHub (pinned to 38031739e5)