In vLLM v0.23.0, `vllm:request_time_per_output_token_seconds` observes one per-finished-request mean decode interval divided by output-token intervals after the first token.

MetricvLLM v0.23.0confidence: highverified 11 hr agoverified-code

PrometheusStatLoggerV1 metrics

vLLM v0.23.0 v1 metrics stats.py(vllm/v1/metrics/loggers.py:439-475)

definesprimarysource_codecode_introspection

Pinned release-tag semantics; do not carry to another version without continuity evidence.

In vLLM v0.23.0, vllm:request_time_per_output_token_seconds observes one per-finished-request mean decode interval divided by output-token intervals after the first token.