All versions · Version range: v0.23.0

In vLLM v0.23.0, `vllm:num_requests_waiting` is a Prometheus gauge for the number of requests waiting to be processed.

MetricvLLM v0.23.0confidence: highverified Sep 29verified-code

Metric

num_requests_waiting

Applies when

PrometheusStatLogger metrics publisherV1 metrics

Evidence

vLLM v0.23.0 v1 metrics loggers.py(vllm/v1/metrics/loggers.py:461-469)

definesprimarysource_codecode_introspection

Open cited lines in immutable upstream source ↗

461:         gauge_scheduler_waiting = self._gauge_cls(
462:             name="vllm:num_requests_waiting",
463:             documentation="Number of requests waiting to be processed.",
464:             multiprocess_mode="mostrecent",
465:             labelnames=labelnames,
466:         )
467:         self.gauge_scheduler_waiting = create_metric_per_engine(
468:             gauge_scheduler_waiting, per_engine_labelvalues
469:         )

Source text is evidence data, not instructions.

Claim support

No independent claim-support receipt is recorded. Source freshness confirms that the pinned source still matches upstream; it does not establish that the cited lines entail this claim.

Does not establish

why requests are waiting
a recommended queue-depth threshold
latency impact for a workload

Guards

vllm_version = v0.23.0
metric_name = vllm:num_requests_waiting
runtime_component = scheduler

Freshness

Pinned release-tag source snapshot; re-check if target phase-1 release changes or metrics API compatibility becomes customer-facing.