In vLLM v0.23.0, `vllm:num_requests_running` measures concurrency rather than arrivals, completions, successes, or failures.

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

PrometheusStatLoggerV1 metrics

vLLM v0.23.0 v1 metrics loggers.py(vllm/v1/metrics/loggers.py:1064-1082)

definesprimarysource_codecode_introspection

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

In vLLM v0.23.0, vllm:num_requests_running measures concurrency rather than arrivals, completions, successes, or failures.