invoke-ai/InvokeAI · error · ValueError

The selected saved workflow must not contain more than one w

Error message

The selected saved workflow must not contain more than one workflow_return node.

What it means

When a workflow-call node executes a saved child workflow, InvokeAI requires exactly one workflow_return node in that child graph so the parent knows which node's output to hand back. This ValueError is raised in get_child_workflow_return_output when the child graph contains more than one workflow_return node, making the return value ambiguous.

Source

Thrown at invokeai/app/services/session_processor/workflow_call_runtime.py:145

    @staticmethod
    def get_waiting_workflow_call_invocation(queue_item: SessionQueueItem) -> CallSavedWorkflowInvocation:
        waiting_frame = queue_item.session.waiting_workflow_call
        if waiting_frame is None:
            raise ValueError("Execution state is not waiting on a workflow call.")
        invocation = queue_item.session.execution_graph.nodes.get(waiting_frame.prepared_call_node_id)
        if not isinstance(invocation, CallSavedWorkflowInvocation):
            raise ValueError("Waiting workflow call frame does not point to a call_saved_workflow invocation.")
        return invocation

    @staticmethod
    def get_child_workflow_return_output(child_session: GraphExecutionState) -> WorkflowReturnOutput:
        workflow_return_node_ids = [
            node_id for node_id, node in child_session.graph.nodes.items() if node.get_type() == "workflow_return"
        ]
        if not workflow_return_node_ids:
            raise ValueError("The selected saved workflow must contain exactly one workflow_return node.")
        if len(workflow_return_node_ids) > 1:
            raise ValueError("The selected saved workflow must not contain more than one workflow_return node.")

        workflow_return_node_id = workflow_return_node_ids[0]
        prepared_return_node_ids = child_session.source_prepared_mapping.get(workflow_return_node_id, set())
        if len(prepared_return_node_ids) != 1:
            raise ValueError(
                "The selected saved workflow produced an unsupported number of workflow_return executions."
            )

        prepared_return_node_id = next(iter(prepared_return_node_ids))
        output = child_session.results.get(prepared_return_node_id)
        if not isinstance(output, WorkflowReturnOutput):
            raise ValueError("The selected saved workflow did not produce a valid workflow_return output.")

        return output

    def resume_waiting_workflow_call(self, queue_item: SessionQueueItem) -> None:
        invocation = self.get_waiting_workflow_call_invocation(queue_item)
        child_session = queue_item.session.waiting_workflow_call_child_session

View on GitHub (pinned to 0b6a024f2f)

Solutions

1. Open the saved child workflow in the editor and delete the extra workflow_return node(s), leaving exactly one
2. Inspect the workflow JSON (graph.nodes) and remove all but one node whose type is workflow_return
3. Re-save the workflow and re-run the parent workflow
4. If the workflow was duplicated programmatically, fix the duplication logic to strip duplicate return nodes

Example fix

// before: graph with two workflow_return nodes
{"nodes": {"a": {"type": "workflow_return", ...}, "b": {"type": "workflow_return", ...}}}
// after: keep exactly one
{"nodes": {"a": {"type": "workflow_return", ...}}}

return_ids = [nid for nid, n in workflow.graph.nodes.items() if n.get_type() == 'workflow_return']
if len(return_ids) != 1:
    raise ValueError(f"Saved workflow must contain exactly one workflow_return node (found {len(return_ids)}).")

def has_single_return(graph) -> bool:
    return sum(1 for n in graph.nodes.values() if n.get_type() == "workflow_return") == 1

try:
    output = runtime.get_child_workflow_return_output(child_session)
except ValueError as e:
    logger.error(f"Child workflow misconfigured: {e}")
    # fix the workflow, then re-run

When it happens

Trigger: Calling a child workflow via a workflow_call invocation whose saved graph was edited (or duplicated) to include two or more workflow_return nodes; the error surfaces when the parent resumes from the completed/waiting child session (resume_waiting_workflow_call or _resume_parent_from_completed_child).

Common situations: Duplicating a workflow that already had a workflow_return node, manually editing a saved workflow JSON, or merging two workflows without removing one of the return nodes.

Related errors

• A saved workflow must be selected before executing call_save
• The selected saved workflow '${self.workflow_id}' could not
• Workflow return key must not be empty.
• Duplicate workflow return key '{key}'.
• Workflow return key '{key}' was not found.

AI-assisted analysis of invoke-ai/InvokeAI@0b6a024f2f (2026-08-29). Data as JSON: /api/errors/eedc2fb7b0618328. Report an issue: GitHub.