`nemo_microservices.types.generation_stats`#

Module Contents#

class nemo_microservices.types.generation_stats.GenerationStats(/, **data: typing.Any)#

Bases: nemo_microservices._models.BaseModel

dialog_rails_duration: Optional[float]#

None

The time in seconds spent in processing the dialog rails.

generation_rails_duration: Optional[float]#

None

The time in seconds spent in generation rails.

input_rails_duration: Optional[float]#

None

The time in seconds spent in processing the input rails.

llm_calls_count: Optional[int]#

None

The number of LLM calls in total.

llm_calls_duration: Optional[float]#

None

The time in seconds spent in LLM calls.

llm_calls_total_completion_tokens: Optional[int]#

None

The total number of completion tokens.

llm_calls_total_prompt_tokens: Optional[int]#

None

The total number of prompt tokens.

llm_calls_total_tokens: Optional[int]#

None

The total number of tokens.

output_rails_duration: Optional[float]#

None

The time in seconds spent in processing the output rails.

total_duration: Optional[float]#

None

The total time in seconds.