xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-04-02 01:27:03 +08:00
Code Issues Packages Projects Releases Wiki Activity

updated

Browse Source 
Signed-off-by: Robert Shaw <robshaw@redhat.com>
This commit is contained in:
Robert Shaw 2025-07-20 17:12:27 +00:00
parent d9291f998e
commit 876c864d3c
1 changed files with 11 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
vllm/v1/metrics/loggers.py
View File

@ -168,8 +168,17 @@ class PrometheusStatLogger(StatLoggerBase):
model_name = vllm_config.model_config.served_model_name
max_model_len = vllm_config.model_config.max_model_len
# self.spec_decoding_prom = self._spec_decoding_cls(
# vllm_config.speculative_config, labelnames, labelvalues)
if len(self.engine_indexes) > 1:
raise NotImplementedError("Prometheus metrics with Spec Decoding "
"with >1 EngineCore per AsyncLLM is not "
"supported yet.")
spec_decode_labelvalues = [
vllm_config.model_config.served_model_name,
str(self.engine_indexes[0])
]
self.spec_decoding_prom = self._spec_decoding_cls(
vllm_config.speculative_config, labelnames,
spec_decode_labelvalues)
#
# Scheduler state