All versions · Version range: v0.23.0

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

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

Metric

estimated_flops_per_gpu_total

Applies when

PrometheusStatLogger metrics publisherV1 metrics

Evidence

vLLM v0.23.0 v1 metrics perf.py(vllm/v1/metrics/perf.py:1286-1293)

definesprimarysource_codecode_introspection

Open cited lines in immutable upstream source ↗

1286:         counter_flops = self._counter_cls(
1287:             name="vllm:estimated_flops_per_gpu_total",
1288:             documentation=(
1289:                 "Estimated number of floating point operations per GPU "
1290:                 "(for Model Flops Utilization calculations)."
1291:             ),
1292:             labelnames=labelnames,
1293:         )

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:estimated_flops_per_gpu_total

Freshness

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