Async check model call limits before making a model call.
abefore_model(
self,
state: ModelCallLimitState[ResponseT],
runtime: Runtime[ContextT]
) -> dict[str, Any] | None| Name | Type | Description |
|---|---|---|
state* | ModelCallLimitState[ResponseT] | The current agent state containing call counts. |
runtime* | Runtime[ContextT] | The langgraph runtime. |