pub async fn generate_streaming(
client: &dyn LlmClient,
req: &StructuredRequest,
on_partial: PartialObjectCallback,
) -> Result<StructuredResult>Expand description
Generate a structured JSON object with streaming partial updates.
Calls on_partial with progressively more complete partial objects as tokens
arrive. Returns the final validated object.
In streaming mode, max_repair_attempts defaults to 0 because a repair
would reset the partial object stream (confusing for consumers).