factor_agent.llm.base.Reply#

class Reply(text, usage, model_id, params=<factory>, raw='')[source]#

Bases: object

The provider’s answer: its text, token usage as the provider reports it and the model that answered.

params are the parameters the backend sent with the prompt (such as max_tokens), and raw is the provider’s reply as it came, before the text was taken out of it.

Attributes#

Reply.raw = ''#

Reply.text#

Reply.usage#

Reply.model_id#

Reply.params#