GenerationStats#

class nemo_microservices.types.GenerationStats(*args: Any, **kwargs: Any)

Bases: BaseModel

dialog_rails_duration: float | None = None: The time in seconds spent in processing the dialog rails.

generation_rails_duration: float | None = None: The time in seconds spent in generation rails.

input_rails_duration: float | None = None: The time in seconds spent in processing the input rails.

llm_calls_count: int | None = None: The number of LLM calls in total.

llm_calls_duration: float | None = None: The time in seconds spent in LLM calls.

llm_calls_total_completion_tokens: int | None = None: The total number of completion tokens.

llm_calls_total_prompt_tokens: int | None = None: The total number of prompt tokens.

llm_calls_total_tokens: int | None = None: The total number of tokens.

output_rails_duration: float | None = None: The time in seconds spent in processing the output rails.

total_duration: float | None = None: The total time in seconds.