biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[V1][Metrics] Add iteration_tokens_total histogram from V0 (#13288)

Browse Source
This commit is contained in:
Mark McLoughlin
2025-02-15 11:56:19 +00:00
committed by GitHub
parent 7fdaaf48ef
commit 2ad1bc7afe
3 changed files with 35 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

13
tests/entrypoints/openai/test_metrics.py
View File

@@ -96,9 +96,14 @@ EXPECTED_VALUES = {
[("_sum", _NUM_REQUESTS * _NUM_GENERATION_TOKENS_PER_REQUEST),
("_count", _NUM_REQUESTS)],
"vllm:request_params_n": [("_count", _NUM_REQUESTS)],
"vllm:request_params_max_tokens":
[("_sum", _NUM_REQUESTS * _NUM_GENERATION_TOKENS_PER_REQUEST),
("_count", _NUM_REQUESTS)],
"vllm:request_params_max_tokens": [
("_sum", _NUM_REQUESTS * _NUM_GENERATION_TOKENS_PER_REQUEST),
("_count", _NUM_REQUESTS)
],
"vllm:iteration_tokens_total":
[("_sum", _NUM_REQUESTS *
(_NUM_PROMPT_TOKENS_PER_REQUEST + _NUM_GENERATION_TOKENS_PER_REQUEST)),
("_count", _NUM_REQUESTS * _NUM_GENERATION_TOKENS_PER_REQUEST)],
"vllm:prompt_tokens": [("_total",
_NUM_REQUESTS * _NUM_PROMPT_TOKENS_PER_REQUEST)],
"vllm:generation_tokens": [
@@ -197,6 +202,7 @@ EXPECTED_METRICS = [
"vllm:request_params_max_tokens_sum",
"vllm:request_params_max_tokens_bucket",
"vllm:request_params_max_tokens_count",
"vllm:iteration_tokens_total",
"vllm:num_preemptions_total",
"vllm:prompt_tokens_total",
"vllm:generation_tokens_total",
@@ -223,6 +229,7 @@ EXPECTED_METRICS_V1 = [
"vllm:gpu_prefix_cache_hits",
"vllm:prompt_tokens_total",
"vllm:generation_tokens_total",
"vllm:iteration_tokens_total",
"vllm:request_success_total",
"vllm:request_prompt_tokens_sum",
"vllm:request_prompt_tokens_bucket",