跳到主要内容

openhands.llm.metrics

ResponseLatency Objects​

class ResponseLatency(BaseModel)

Metric tracking the round-trip time per completion call.

TokenUsage Objects​

class TokenUsage(BaseModel)

Metric tracking detailed token usage per completion call.

Metrics Objects​

class Metrics()

Metrics class can record various metrics during running and evaluation. We track:

  • accumulated_cost and costs
  • A list of ResponseLatency
  • A list of TokenUsage (one per call).

add_token_usage​

def add_token_usage(prompt_tokens: int, completion_tokens: int,
                    cache_read_tokens: int, cache_write_tokens: int,
                    response_id: str) -> None

Add a single usage record.

merge​

def merge(other: 'Metrics') -> None

Merge 'other' metrics into this one.

get​

def get() -> dict

Return the metrics in a dictionary.

log​

def log()

Log the metrics.