vLLM v0.23.0 · Version scoped

In vLLM v0.23.0, `vllm:mm_cache_queries` is registered as a Prometheus counter.

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

Metric

mm_cache_queries

Applies when

PrometheusStatLogger metrics publisherV1 metrics

Evidence

vLLM v0.23.0 v1 metrics loggers.py(vllm/v1/metrics/loggers.py:594-600)

definesprimarysource_codecode_introspection

Open cited lines in immutable upstream source ↗

594:         counter_mm_cache_queries = self._counter_cls(
595:             name="vllm:mm_cache_queries",
596:             documentation=(
597:                 "Multi-modal cache queries, in terms of number of queried items."
598:             ),
599:             labelnames=labelnames,
600:         )

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

the metric's unit of measure
what the metric measures semantically
a healthy or expected value range

Guards

vllm_version = v0.23.0
metric_name = vllm:mm_cache_queries

Freshness

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