# VLLM Engine Metrics

**URL:** <https://discuss.vllm.ai/t/vllm-engine-metrics/810>\
**Category:** Benchmarking\
**Created:** [June 11, 2025, 7:48pm UTC](https://discuss.vllm.ai/t/vllm-engine-metrics/810 "2025-06-11T19:48:03Z")\
**Posts on this page:** 1\
**Page:** 2

<div class="post-metadata">

**Author:** ![RunLLM](https://yyz2.discourse-cdn.com/flex036/user_avatar/discuss.vllm.ai/runllm/32/233_2.png) [@RunLLM](https://discuss.vllm.ai/u/RunLLM)\
**Post date:** [June 11, 2025, 11:18pm UTC](https://discuss.vllm.ai/t/vllm-engine-metrics/810/21 "2025-06-11T23:18:44Z")

</div>

Yes, your code correctly enables vLLM metrics, including the histograms vllm:time\_to\_first\_token\_seconds, vllm:time\_per\_output\_token\_seconds, and vllm:e2e\_request\_latency\_seconds, by instantiating both LoggingStatLogger and PrometheusStatLogger in the engine. These loggers will collect and expose the metrics you listed, as described in the [metrics documentation](https://docs.vllm.ai/en/latest/design/v1/metrics.html) and [metrics API reference](https://docs.vllm.ai/en/latest/api/vllm/vllm.engine.metrics.html).

Just ensure the PrometheusStatLogger is active and the /metrics endpoint is exposed by your server for Prometheus to scrape the data.

Would you like more detail on how to verify these metrics are being collected?

Sources:

- [metrics documentation](https://docs.vllm.ai/en/latest/design/v1/metrics.html)
- [metrics API reference](https://docs.vllm.ai/en/latest/api/vllm/vllm.engine.metrics.html)

[Previous page](https://discuss.vllm.ai/t/vllm-engine-metrics/810.md?page=1)