[V1][Metrics] Add per-request prompt/generation_tokens histograms (#12516)

Signed-off-by: Mark McLoughlin <markmc@redhat.com>
2025-01-28 22:07:22 +00:00
parent f26d790718
commit c386c43ca3
5 changed files with 103 additions and 15 deletions
--- a/vllm/v1/engine/async_llm.py
+++ b/vllm/v1/engine/async_llm.py
@@ -53,8 +53,7 @@ class AsyncLLM(EngineClient):
        self.log_stats = log_stats
        self.stat_loggers: List[StatLoggerBase] = [
            LoggingStatLogger(),
-            PrometheusStatLogger(labels=dict(
-                model_name=self.model_config.served_model_name)),
+            PrometheusStatLogger(vllm_config.model_config),
        ]

        # Tokenizer (+ ensure liveness if running in another process).