vllm:request_max_num_generation_tokens
Histogram of the maximum requested generation tokens per request.
MetricPrometheusvLLM v0.23.0verified Jun 25
Claims (0)
Availability
vllm:request_max_num_generation_tokens is registered as a Prometheus histogram in vLLM v0.23.0.vllm.surface.prometheus_metricscomplete_inventoryPrometheusStatLogger metrics publisher