metrics: support customer buckets for prompt/generation_tokens_histogram (#9634)

This commit is contained in:
Yingchun Lai
2025-09-04 22:22:08 +08:00
committed by GitHub
parent 75ee00112d
commit b32ab0705e
7 changed files with 293 additions and 19 deletions

View File

@@ -329,6 +329,7 @@ class TokenizerManager:
# Metrics
if self.enable_metrics:
self.metrics_collector = TokenizerMetricsCollector(
server_args=server_args,
labels={
"model_name": self.server_args.served_model_name,
# TODO: Add lora name/path in the future,