[frontend] add selector for median or p99 for e2el, interactivity, and ttft / [frontend] 为 e2el、交互性和 ttft 添加中位数或 p99 选择器
@adibarra is already working on this.
Since Feb 19, 2026.
Assessment
This issue has not been assessed yet.
Description
Right now, we use the median data. P99 data data also exists for each of these data points. For instance, take the following artifact:
{
"hw": "gb200",
"conc": 4,
"image": "lmsysorg/sglang:v0.5.8-cu130",
"model": "deepseek-ai/DeepSeek-R1",
"infmax_model_prefix": "dsr1",
"framework": "dynamo-sglang",
"precision": "fp4",
"spec_decoding": "none",
"disagg": true,
"isl": 8192,
"osl": 1024,
"is_multinode": true,
"prefill_tp": 4,
"prefill_ep": 1,
"prefill_dp_attention": "false",
"prefill_num_workers": 1,
"decode_tp": 4,
"decode_ep": 1,
"decode_dp_attention": "false",
"decode_num_workers": 4,
"num_prefill_gpu": 4,
"num_decode_gpu": 16,
"tput_per_gpu": 239.08275342140536,
"output_tput_per_gpu": 33.241061135990364,
"input_tput_per_gpu": 1062.4495225630653,
"mean_ttft": 0.43425333057239185,
"median_ttft": 0.25317291749524884,
"std_ttft": 0.36127142112833116,
"p99_ttft": 1.48367712290521,
"mean_tpot": 0.006977059080877132,
"mean_intvty": 143.32686428595977,
"median_tpot": 0.006802988651651399,
"median_intvty": 146.9942184538635,
"std_tpot": 0.0005388865723481444,
"std_intvty": 1855.6780801618415,
"p99_tpot": 0.008094592967696608,
"p99_intvty": 123.53925688304984,
"mean_itl": 0.34069569860559523,
"median_itl": 0.33583703998010606,
"std_itl": 0.06460103068039673,
"p99_itl": 0.39620181820704603,
"mean_e2el": 6.8478497144227735,
"median_e2el": 6.729302968495176,
"std_e2el": 0.8481532815896575,
"p99_e2el": 8.62211772535491
}
Users should have the option to select between SLA (median or P99 for now). P99 should be the default.
中文说明
前端为 e2el、交互性和 TTFT 添加中位数或 p99 选择器。
- Dominant language
- Python
- Stars
- 1.7k
- Forks
- 303
- Avg merge
- 1d 13h
- Merged PRs (30d)
- 284
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from SemiAnalysisAI/InferenceX
-
Difficulty 2/5 1-3 hours Newbie friendliness 68/100
SemiAnalysisAI/InferenceX#2125 ·
-
Difficulty 2/5 1-3 hours Newbie friendliness 68/100
SemiAnalysisAI/InferenceX#1587 ·
-
Difficulty 1/5 Under an hour Newbie friendliness 78/100
SemiAnalysisAI/InferenceX#1369 · 3 comments ·
-
Difficulty 1/5 1-3 hours Newbie friendliness 76/100
SemiAnalysisAI/InferenceX#1359 · 1 comment ·
-
Difficulty 5/5 Over a week Newbie friendliness 30/100
SemiAnalysisAI/InferenceX#3122 · 3 comments ·
All issues in SemiAnalysisAI/InferenceX
Similar issues
-
Difficulty 2/5 1-3 hours Newbie friendliness 74/100
bancolombia/sentinel#23 ·
-
test md OpenCI
Difficulty 2/5 1-3 hours Newbie friendliness 74/100
-
integration:quickjs org:external priority:backlog topic:code-interpreter topic:middleware type:feature
Difficulty 2/5 1-3 hours Newbie friendliness 74/100
langchain-ai/deepagents#6450 ·
-
bug client
Difficulty 2/5 1-3 hours Newbie friendliness 88/100
-
Difficulty 2/5 1-3 hours Newbie friendliness 74/100