chatWithProgress

Returns the model's next message for request, like ChatModel.chat, and lets a streamed run watch the model write it. Use it in a node in place of ChatModel.chat:

val answer = node("answer", work = { email -> model.chatWithProgress(ChatRequest(email.messages)).message }) { email, reply ->
    email.copy(messages = email.messages + reply)
}

graph.stream(email).collect { event -> event.textDelta?.let(::print) }

In a node of a run that is collected with stream or streamResume, the model is asked with ChatModel.stream and each ChatEvent.TextDelta is passed to reportProgress, so it arrives in the run's flow as a GraphEvent.NodeProgress; textDelta reads it there. Anywhere else nobody would see the pieces, and the model is asked with ChatModel.chat.

Throws

Sends prompt as a single user message and returns the text of the answer, like chat with a prompt, and lets a streamed run watch the model write it. See the overload with a ChatRequest.

node("summarize", work = { model.chatWithProgress("Summarize: ${it.notes}") }) { state, summary ->
    state.copy(summary = summary)
}