[Bug]: avg_generation_throughput_toks_per_s missing

[zbk2012]

Your current environment

After upgrading vllm 0.5.3. post1 to vllm 0.7.2, the avg.generation_throughput_toks_per_2 monitoring metric is missing from the metrics.

🐛 Describe the bug

After upgrading vllm 0.5.3. post1 to vllm 0.7.2, the avg.generation_throughput_toks_per_2 monitoring metric is missing from the metrics.

Before submitting a new issue...

• Make sure you already searched for relevant issues, and asked the chatbot living at the bottom right corner of the documentation page, which can answer lots of frequently asked questions.

[markmc]

Thank you for your interest in vLLM.

This metric was deprecated by #2764 and removed in #12383. You should use vllm:generation_tokens_total counter instead.

[tkg61]

@markmc do you have a simple example on how this should be done since totals is different than a live view of the rate per second?

[markmc]

Try using the rate() function in Prometheus - for example, rate(vllm:generation_tokens_total[10m]). This is what our example Grafana dashboard does.

[zbk2012]

ok，thank you。

[github-actions]

This issue has been automatically marked as stale because it has not had any activity within 90 days. It will be automatically closed if no further activity occurs within 30 days. Leave a comment if you feel this issue should remain open. Thank you!

[FelipePavan]

The solution provided by @markmc works, you guys can close this issue