Ask a question to get started
Enter to send•Shift+Enter new line
after_model( self, state: CostState, runtime: Runtime[ContextT] ) -> dict
state
CostState
runtime
Runtime[ContextT]
Charge every model request completed since the previous checkpoint.
Current state containing messages and prior session cost.
LangGraph runtime used to identify the thread and to stream the new total to the client.
