ModelResponse
A complete model response, as a found lookup returns it: the output parts, why it stopped, token usage and the provider's request id.
Fields
contentreadonly OutputPart[]requiredThe output parts in order: text, tool calls, reasoning and hosted tool results.
stop_reasonModelResponseEvent.data.stop_reasonrequiredOnly end_turn, stop_sequence and refusal complete a turn; tool_use dispatches; max_tokens continues; pause_turn re-requests with the paused content as-is, up to context.max_pause_continuations; context_window_exceeded ends the turn context_exhausted; other (a reason the adapter can't name) ends it error.
usageUsagerequiredPer-field measured-or-unknown usage.
provider_request_idNonEmpty | nullrequiredThe provider's id for this request, or null when it gives none. A found model lookup must carry it, since model_response_recovered records it.
ModelRequest
One send attempt: the client request id and the exact rendered request bytes the adapter encodes for its provider.
ModelChunk
One item of a Model.send stream: a text delta, a finished output part, the final done chunk, or one terminal rejected chunk when the send failed before any content.