Files
aituner/runs/frontier-attn-structured-v0/results/trace-verdict.json

2219 lines
67 KiB
JSON

{
"schema": "frontier-attn-structured-trial-aware-verdict-v1",
"metric_note": "Primary values are per-real-trial distribution biases with request alignment by index. legacy_pooled reproduces the old milestone quantile convention only for direct comparison.",
"cells": {
"tp1_rho0p00125": {
"old": {
"trialwise_distribution_bias": [
{
"ttft": {
"mean": -0.13148990670086547,
"p50": -0.18680026171265385,
"p90": -0.2281339774144484,
"p99": -0.14956660473341468
},
"tpot": {
"mean": 0.1393089430926731,
"p50": 0.17688566049293195,
"p90": 0.021918816469380976,
"p99": -0.06745527151165458
},
"e2e": {
"mean": 0.0728867501966229,
"p50": 0.11334615386654034,
"p90": 0.11968616309773676,
"p99": -0.04061861992039894
}
},
{
"ttft": {
"mean": -0.13772594464937732,
"p50": -0.19220957147090617,
"p90": -0.2342597777065415,
"p99": -0.15374940938906825
},
"tpot": {
"mean": 0.12195501379179413,
"p50": 0.1610213245142966,
"p90": 0.005473527596398301,
"p99": -0.08127351500612069
},
"e2e": {
"mean": 0.05800039446459018,
"p50": 0.09970261087805131,
"p90": 0.10532164717109044,
"p99": -0.046957770855300444
}
}
],
"trialwise_distribution_bias_summary": {
"ttft": {
"mean": {
"mean": -0.1346079256751214,
"min": -0.13772594464937732,
"max": -0.13148990670086547
},
"p50": {
"mean": -0.18950491659178,
"min": -0.19220957147090617,
"max": -0.18680026171265385
},
"p90": {
"mean": -0.23119687756049495,
"min": -0.2342597777065415,
"max": -0.2281339774144484
},
"p99": {
"mean": -0.15165800706124147,
"min": -0.15374940938906825,
"max": -0.14956660473341468
}
},
"tpot": {
"mean": {
"mean": 0.13063197844223362,
"min": 0.12195501379179413,
"max": 0.1393089430926731
},
"p50": {
"mean": 0.16895349250361427,
"min": 0.1610213245142966,
"max": 0.17688566049293195
},
"p90": {
"mean": 0.013696172032889638,
"min": 0.005473527596398301,
"max": 0.021918816469380976
},
"p99": {
"mean": -0.07436439325888763,
"min": -0.08127351500612069,
"max": -0.06745527151165458
}
},
"e2e": {
"mean": {
"mean": 0.06544357233060655,
"min": 0.05800039446459018,
"max": 0.0728867501966229
},
"p50": {
"mean": 0.10652438237229583,
"min": 0.09970261087805131,
"max": 0.11334615386654034
},
"p90": {
"mean": 0.1125039051344136,
"min": 0.10532164717109044,
"max": 0.11968616309773676
},
"p99": {
"mean": -0.04378819538784969,
"min": -0.046957770855300444,
"max": -0.04061861992039894
}
}
},
"legacy_pooled_distribution_bias": {
"ttft": {
"mean": -0.13461915993830945,
"p50": -0.1888032883165517,
"p90": -0.23223319000679374,
"p99": -0.16824856840168442
},
"tpot": {
"mean": 0.1305653876178269,
"p50": 0.16837991778063455,
"p90": 0.012855564422932954,
"p99": -0.08151080694091946
},
"e2e": {
"mean": 0.06539157436639341,
"p50": 0.10867934171755954,
"p90": 0.10946179160934073,
"p99": -0.0657146472433028
}
}
},
"new": {
"trialwise_distribution_bias": [
{
"ttft": {
"mean": -0.05989664674711635,
"p50": -0.10957660686565794,
"p90": -0.0836059887499959,
"p99": -0.0653231144095356
},
"tpot": {
"mean": 0.1518790062372861,
"p50": 0.17738738332864315,
"p90": 0.032091888121311134,
"p99": -0.05049862983288826
},
"e2e": {
"mean": 0.09143423903966252,
"p50": 0.11610023742963112,
"p90": 0.12903581929912153,
"p99": -0.020226996786041312
}
},
{
"ttft": {
"mean": -0.06664673546990688,
"p50": -0.11549959936391153,
"p90": -0.09087881400402557,
"p99": -0.0699202655213567
},
"tpot": {
"mean": 0.1343336100051224,
"p50": 0.16151628416144917,
"p90": 0.015482888492302317,
"p99": -0.0645681331290643
},
"e2e": {
"mean": 0.07629053599972169,
"p50": 0.10242294441887236,
"p90": 0.11455135611418107,
"p99": -0.02670088618839733
}
}
],
"trialwise_distribution_bias_summary": {
"ttft": {
"mean": {
"mean": -0.06327169110851161,
"min": -0.06664673546990688,
"max": -0.05989664674711635
},
"p50": {
"mean": -0.11253810311478474,
"min": -0.11549959936391153,
"max": -0.10957660686565794
},
"p90": {
"mean": -0.08724240137701074,
"min": -0.09087881400402557,
"max": -0.0836059887499959
},
"p99": {
"mean": -0.06762168996544615,
"min": -0.0699202655213567,
"max": -0.0653231144095356
}
},
"tpot": {
"mean": {
"mean": 0.14310630812120423,
"min": 0.1343336100051224,
"max": 0.1518790062372861
},
"p50": {
"mean": 0.16945183374504616,
"min": 0.16151628416144917,
"max": 0.17738738332864315
},
"p90": {
"mean": 0.023787388306806725,
"min": 0.015482888492302317,
"max": 0.032091888121311134
},
"p99": {
"mean": -0.05753338148097628,
"min": -0.0645681331290643,
"max": -0.05049862983288826
}
},
"e2e": {
"mean": {
"mean": 0.0838623875196921,
"min": 0.07629053599972169,
"max": 0.09143423903966252
},
"p50": {
"mean": 0.10926159092425174,
"min": 0.10242294441887236,
"max": 0.11610023742963112
},
"p90": {
"mean": 0.1217935877066513,
"min": 0.11455135611418107,
"max": 0.12903581929912153
},
"p99": {
"mean": -0.023463941487219322,
"min": -0.02670088618839733,
"max": -0.020226996786041312
}
}
},
"paired_relative_error": [
{
"ttft": {
"mean": 0.3691695743143199,
"p50": -0.03099707280100344,
"p90": 1.7891828562343537,
"p99": 5.429960617790179
},
"tpot": {
"mean": 0.19492722441928262,
"p50": 0.15055142720578293,
"p90": 0.2670903700878269,
"p99": 0.6745762204781113
},
"e2e": {
"mean": 0.1844268367659844,
"p50": 0.1424666323844526,
"p90": 0.31842146555728224,
"p99": 0.8437286984487435
}
},
{
"ttft": {
"mean": 0.3669522667824523,
"p50": -0.03476798007471665,
"p90": 1.827734692042949,
"p99": 5.512577837970861
},
"tpot": {
"mean": 0.1783718677969176,
"p50": 0.13458290148059987,
"p90": 0.24952928603370794,
"p99": 0.6582817886518392
},
"e2e": {
"mean": 0.16858938086274747,
"p50": 0.12880754276426226,
"p90": 0.29822488192148,
"p99": 0.7895217968914289
}
}
],
"legacy_pooled_distribution_bias": {
"ttft": {
"mean": -0.06328385143753473,
"p50": -0.1117698463137391,
"p90": -0.0884727580604037,
"p99": -0.08585570369326061
},
"tpot": {
"mean": 0.14303898259628217,
"p50": 0.1688780144991309,
"p90": 0.02293841255516459,
"p99": -0.06480973978913974
},
"e2e": {
"mean": 0.0838094906426525,
"p50": 0.11142188099086506,
"p90": 0.11872607178176105,
"p99": -0.04585643943465822
}
},
"waiting_p99_ms": 1337.0359972076221,
"validity": "GATE_FAIL_DIAGNOSTIC"
}
},
"tp1_rho0p0025": {
"old": {
"trialwise_distribution_bias": [
{
"ttft": {
"mean": -0.178961187154444,
"p50": 0.004736647098298533,
"p90": -0.26795354312466707,
"p99": -0.22168753840926173
},
"tpot": {
"mean": 0.01036398277932227,
"p50": 0.13356090363025383,
"p90": -0.05541984044976886,
"p99": -0.26072314328609314
},
"e2e": {
"mean": -0.024645268935464575,
"p50": 0.0817280529155681,
"p90": 0.04955125280528564,
"p99": -0.1871016978849053
}
},
{
"ttft": {
"mean": -0.17456884440992257,
"p50": 0.011041256441842635,
"p90": -0.25172604669420906,
"p99": -0.21630340228356143
},
"tpot": {
"mean": 0.01842954963685723,
"p50": 0.13346201996043702,
"p90": -0.05215064358043298,
"p99": -0.15675916344669943
},
"e2e": {
"mean": -0.015641746084433396,
"p50": 0.07195840871873896,
"p90": 0.052264109329363505,
"p99": -0.17396512464775202
}
}
],
"trialwise_distribution_bias_summary": {
"ttft": {
"mean": {
"mean": -0.1767650157821833,
"min": -0.178961187154444,
"max": -0.17456884440992257
},
"p50": {
"mean": 0.007888951770070584,
"min": 0.004736647098298533,
"max": 0.011041256441842635
},
"p90": {
"mean": -0.2598397949094381,
"min": -0.26795354312466707,
"max": -0.25172604669420906
},
"p99": {
"mean": -0.2189954703464116,
"min": -0.22168753840926173,
"max": -0.21630340228356143
}
},
"tpot": {
"mean": {
"mean": 0.01439676620808975,
"min": 0.01036398277932227,
"max": 0.01842954963685723
},
"p50": {
"mean": 0.13351146179534543,
"min": 0.13346201996043702,
"max": 0.13356090363025383
},
"p90": {
"mean": -0.05378524201510092,
"min": -0.05541984044976886,
"max": -0.05215064358043298
},
"p99": {
"mean": -0.2087411533663963,
"min": -0.26072314328609314,
"max": -0.15675916344669943
}
},
"e2e": {
"mean": {
"mean": -0.020143507509948984,
"min": -0.024645268935464575,
"max": -0.015641746084433396
},
"p50": {
"mean": 0.07684323081715352,
"min": 0.07195840871873896,
"max": 0.0817280529155681
},
"p90": {
"mean": 0.05090768106732457,
"min": 0.04955125280528564,
"max": 0.052264109329363505
},
"p99": {
"mean": -0.18053341126632866,
"min": -0.1871016978849053,
"max": -0.17396512464775202
}
}
},
"legacy_pooled_distribution_bias": {
"ttft": {
"mean": -0.17677087458166194,
"p50": 0.008311436147272566,
"p90": -0.2608553178054708,
"p99": -0.22116442296103195
},
"tpot": {
"mean": 0.014380733682430142,
"p50": 0.13351145963877706,
"p90": -0.05456950130438105,
"p99": -0.23635406197836167
},
"e2e": {
"mean": -0.020164189983441452,
"p50": 0.07632815851795742,
"p90": 0.049809009042946335,
"p99": -0.18303183751478602
}
}
},
"new": {
"trialwise_distribution_bias": [
{
"ttft": {
"mean": -0.09728726150141453,
"p50": 0.011908330017980539,
"p90": -0.1382218037906623,
"p99": -0.14467638598472385
},
"tpot": {
"mean": 0.04813594501063553,
"p50": 0.1424913754734047,
"p90": 0.01715387922135076,
"p99": -0.2052193535460428
},
"e2e": {
"mean": 0.010994374852847301,
"p50": 0.11060355438719509,
"p90": 0.08139049920593726,
"p99": -0.15755189584478857
}
},
{
"ttft": {
"mean": -0.09245798463253013,
"p50": 0.018257940864430843,
"p90": -0.11911850444197748,
"p99": -0.13875950941310633
},
"tpot": {
"mean": 0.05650303913152255,
"p50": 0.14239171277369528,
"p90": 0.020674254113796777,
"p99": -0.0934499151356964
},
"e2e": {
"mean": 0.02032688810811885,
"p50": 0.10057312063737019,
"p90": 0.0841856530995708,
"p99": -0.1439377928383722
}
}
],
"trialwise_distribution_bias_summary": {
"ttft": {
"mean": {
"mean": -0.09487262306697233,
"min": -0.09728726150141453,
"max": -0.09245798463253013
},
"p50": {
"mean": 0.015083135441205691,
"min": 0.011908330017980539,
"max": 0.018257940864430843
},
"p90": {
"mean": -0.12867015411631988,
"min": -0.1382218037906623,
"max": -0.11911850444197748
},
"p99": {
"mean": -0.1417179476989151,
"min": -0.14467638598472385,
"max": -0.13875950941310633
}
},
"tpot": {
"mean": {
"mean": 0.05231949207107904,
"min": 0.04813594501063553,
"max": 0.05650303913152255
},
"p50": {
"mean": 0.14244154412354998,
"min": 0.14239171277369528,
"max": 0.1424913754734047
},
"p90": {
"mean": 0.018914066667573767,
"min": 0.01715387922135076,
"max": 0.020674254113796777
},
"p99": {
"mean": -0.14933463434086958,
"min": -0.2052193535460428,
"max": -0.0934499151356964
}
},
"e2e": {
"mean": {
"mean": 0.015660631480483075,
"min": 0.010994374852847301,
"max": 0.02032688810811885
},
"p50": {
"mean": 0.10558833751228264,
"min": 0.10057312063737019,
"max": 0.11060355438719509
},
"p90": {
"mean": 0.08278807615275402,
"min": 0.08139049920593726,
"max": 0.0841856530995708
},
"p99": {
"mean": -0.1507448443415804,
"min": -0.15755189584478857,
"max": -0.1439377928383722
}
}
},
"paired_relative_error": [
{
"ttft": {
"mean": 0.39789106943830893,
"p50": -0.037298357008305895,
"p90": 1.9730972451814688,
"p99": 6.018295451307029
},
"tpot": {
"mean": 0.1553028513608087,
"p50": 0.12385990892916658,
"p90": 0.2584968822537069,
"p99": 0.9658213367115374
},
"e2e": {
"mean": 0.1471398515879261,
"p50": 0.10601983073958936,
"p90": 0.32930377247615056,
"p99": 1.075635524765205
}
},
{
"ttft": {
"mean": 0.4002572158008619,
"p50": -0.031486174528958244,
"p90": 1.8666003636442456,
"p99": 5.991354563261557
},
"tpot": {
"mean": 0.1566771540679317,
"p50": 0.12379976526368523,
"p90": 0.25274583367589193,
"p99": 0.958693609453715
},
"e2e": {
"mean": 0.14911603622614725,
"p50": 0.10650308347135491,
"p90": 0.3195494670044913,
"p99": 1.1185032164459736
}
}
],
"legacy_pooled_distribution_bias": {
"ttft": {
"mean": -0.09487906467830536,
"p50": 0.015508635458377175,
"p90": -0.129865645852306,
"p99": -0.14410151018893788
},
"tpot": {
"mean": 0.052302860177306544,
"p50": 0.14244154194999165,
"p90": 0.018069551387063856,
"p99": -0.1790206784227079
},
"e2e": {
"mean": 0.01563919326555167,
"p50": 0.10505951594320918,
"p90": 0.08165607475145953,
"p99": -0.15333409130015813
}
},
"waiting_p99_ms": 1892.1769691703946,
"validity": "GATE_FAIL_DIAGNOSTIC"
}
},
"tp2_rho0p0025": {
"old": {
"trialwise_distribution_bias": [
{
"ttft": {
"mean": -0.11106335941203382,
"p50": -0.1855202869158692,
"p90": -0.21828233796243518,
"p99": -0.1688334681304367
},
"tpot": {
"mean": 0.1303208667385706,
"p50": 0.16909257797747365,
"p90": 0.05206789287995917,
"p99": 0.14191216788372232
},
"e2e": {
"mean": 0.1031266658038281,
"p50": 0.13020268054402595,
"p90": 0.09830535586040931,
"p99": -0.0581273596953024
}
},
{
"ttft": {
"mean": -0.11215904539344854,
"p50": -0.18289416715548218,
"p90": -0.21914116973979564,
"p99": -0.17209787391954623
},
"tpot": {
"mean": 0.14399128493700397,
"p50": 0.1866331158045056,
"p90": 0.06642456306402275,
"p99": 0.15612928840488144
},
"e2e": {
"mean": 0.11544885011206724,
"p50": 0.14502095370348014,
"p90": 0.10849722315778151,
"p99": -0.045303035491763635
}
}
],
"trialwise_distribution_bias_summary": {
"ttft": {
"mean": {
"mean": -0.11161120240274118,
"min": -0.11215904539344854,
"max": -0.11106335941203382
},
"p50": {
"mean": -0.18420722703567569,
"min": -0.1855202869158692,
"max": -0.18289416715548218
},
"p90": {
"mean": -0.2187117538511154,
"min": -0.21914116973979564,
"max": -0.21828233796243518
},
"p99": {
"mean": -0.17046567102499147,
"min": -0.17209787391954623,
"max": -0.1688334681304367
}
},
"tpot": {
"mean": {
"mean": 0.13715607583778727,
"min": 0.1303208667385706,
"max": 0.14399128493700397
},
"p50": {
"mean": 0.17786284689098963,
"min": 0.16909257797747365,
"max": 0.1866331158045056
},
"p90": {
"mean": 0.05924622797199096,
"min": 0.05206789287995917,
"max": 0.06642456306402275
},
"p99": {
"mean": 0.1490207281443019,
"min": 0.14191216788372232,
"max": 0.15612928840488144
}
},
"e2e": {
"mean": {
"mean": 0.10928775795794766,
"min": 0.1031266658038281,
"max": 0.11544885011206724
},
"p50": {
"mean": 0.13761181712375303,
"min": 0.13020268054402595,
"max": 0.14502095370348014
},
"p90": {
"mean": 0.10340128950909541,
"min": 0.09830535586040931,
"max": 0.10849722315778151
},
"p99": {
"mean": -0.05171519759353302,
"min": -0.0581273596953024,
"max": -0.045303035491763635
}
}
},
"legacy_pooled_distribution_bias": {
"ttft": {
"mean": -0.11161154024124531,
"p50": -0.18420934047220774,
"p90": -0.21836264490202395,
"p99": -0.17283197594971011
},
"tpot": {
"mean": 0.13711499081487563,
"p50": 0.17555321305308488,
"p90": 0.0591579975137338,
"p99": 0.1300483675091633
},
"e2e": {
"mean": 0.10925353865257875,
"p50": 0.1339165600755408,
"p90": 0.10666828724664539,
"p99": -0.07437942975605877
}
}
},
"new": {
"trialwise_distribution_bias": [
{
"ttft": {
"mean": -0.044844475267623835,
"p50": -0.08906525404741798,
"p90": -0.11412907883561804,
"p99": -0.07227786537885415
},
"tpot": {
"mean": 0.14386158526258666,
"p50": 0.1787588067010671,
"p90": 0.071419448886029,
"p99": 0.17844287277146695
},
"e2e": {
"mean": 0.12073678040141796,
"p50": 0.14420262806746403,
"p90": 0.11915070375608593,
"p99": -0.022788557992646938
}
},
{
"ttft": {
"mean": -0.046021781355297144,
"p50": -0.08612813517467069,
"p90": -0.11510233828040589,
"p99": -0.07592149320892366
},
"tpot": {
"mean": 0.15769576871598226,
"p50": 0.196444372264794,
"p90": 0.08604019319402298,
"p99": 0.19311481061458743
},
"e2e": {
"mean": 0.13325567383154438,
"p50": 0.15920445684060733,
"p90": 0.12953600816854335,
"p99": -0.009483068681846096
}
}
],
"trialwise_distribution_bias_summary": {
"ttft": {
"mean": {
"mean": -0.045433128311460486,
"min": -0.046021781355297144,
"max": -0.044844475267623835
},
"p50": {
"mean": -0.08759669461104433,
"min": -0.08906525404741798,
"max": -0.08612813517467069
},
"p90": {
"mean": -0.11461570855801197,
"min": -0.11510233828040589,
"max": -0.11412907883561804
},
"p99": {
"mean": -0.07409967929388891,
"min": -0.07592149320892366,
"max": -0.07227786537885415
}
},
"tpot": {
"mean": {
"mean": 0.15077867698928446,
"min": 0.14386158526258666,
"max": 0.15769576871598226
},
"p50": {
"mean": 0.18760158948293054,
"min": 0.1787588067010671,
"max": 0.196444372264794
},
"p90": {
"mean": 0.07872982104002599,
"min": 0.071419448886029,
"max": 0.08604019319402298
},
"p99": {
"mean": 0.1857788416930272,
"min": 0.17844287277146695,
"max": 0.19311481061458743
}
},
"e2e": {
"mean": {
"mean": 0.12699622711648118,
"min": 0.12073678040141796,
"max": 0.13325567383154438
},
"p50": {
"mean": 0.15170354245403567,
"min": 0.14420262806746403,
"max": 0.15920445684060733
},
"p90": {
"mean": 0.12434335596231463,
"min": 0.11915070375608593,
"max": 0.12953600816854335
},
"p99": {
"mean": -0.016135813337246518,
"min": -0.022788557992646938,
"max": -0.009483068681846096
}
}
},
"paired_relative_error": [
{
"ttft": {
"mean": 0.28317282752846445,
"p50": 0.0015652470674283874,
"p90": 0.7651897154483899,
"p99": 4.595783098224935
},
"tpot": {
"mean": 0.1752842205127379,
"p50": 0.17053139492828456,
"p90": 0.24211599227572878,
"p99": 0.7603540706787565
},
"e2e": {
"mean": 0.16397230912194014,
"p50": 0.14764836952252053,
"p90": 0.26575788904393577,
"p99": 0.6911771461107379
}
},
{
"ttft": {
"mean": 0.27885319671790226,
"p50": 0.0008797795879422852,
"p90": 0.7521533350007815,
"p99": 4.589795377891504
},
"tpot": {
"mean": 0.18903026397070546,
"p50": 0.18546796656470424,
"p90": 0.25643004508741135,
"p99": 0.7281958169280102
},
"e2e": {
"mean": 0.17634926693766606,
"p50": 0.15964733210774384,
"p90": 0.2801020190882037,
"p99": 0.7024304395510836
}
}
],
"legacy_pooled_distribution_bias": {
"ttft": {
"mean": -0.045433491316312524,
"p50": -0.0875990583320057,
"p90": -0.1142200855839675,
"p99": -0.07674087497700505
},
"tpot": {
"mean": 0.15073709978689778,
"p50": 0.18527285925405948,
"p90": 0.07863996768588354,
"p99": 0.1661995397125918
},
"e2e": {
"mean": 0.12696146154030977,
"p50": 0.14796251184179712,
"p90": 0.12767235986604622,
"p99": -0.03965040123988098
}
},
"waiting_p99_ms": 777.5911879793948,
"validity": "PASS_SUBCRITICAL"
}
},
"tp2_rho0p005": {
"old": {
"trialwise_distribution_bias": [
{
"ttft": {
"mean": -0.141581861927082,
"p50": -0.2051822104110931,
"p90": -0.1895782660551383,
"p99": -0.19346662831715172
},
"tpot": {
"mean": 0.025318500347963065,
"p50": 0.07182548360524435,
"p90": -0.04507556167777899,
"p99": -0.10054590154610009
},
"e2e": {
"mean": 0.01846906077214167,
"p50": 0.07205509683921338,
"p90": -0.037157045612690995,
"p99": -0.051216767638022506
}
},
{
"ttft": {
"mean": -0.1410153906245514,
"p50": -0.20356790487770302,
"p90": -0.18910363664562158,
"p99": -0.192398015096606
},
"tpot": {
"mean": 0.03709549859113398,
"p50": 0.08312549432043648,
"p90": -0.029759535186826266,
"p99": -0.09947054412002211
},
"e2e": {
"mean": 0.029324095335773688,
"p50": 0.07587484709261458,
"p90": -0.027431862141837213,
"p99": -0.04314368195514718
}
}
],
"trialwise_distribution_bias_summary": {
"ttft": {
"mean": {
"mean": -0.14129862627581669,
"min": -0.141581861927082,
"max": -0.1410153906245514
},
"p50": {
"mean": -0.20437505764439806,
"min": -0.2051822104110931,
"max": -0.20356790487770302
},
"p90": {
"mean": -0.18934095135037993,
"min": -0.1895782660551383,
"max": -0.18910363664562158
},
"p99": {
"mean": -0.19293232170687885,
"min": -0.19346662831715172,
"max": -0.192398015096606
}
},
"tpot": {
"mean": {
"mean": 0.031206999469548524,
"min": 0.025318500347963065,
"max": 0.03709549859113398
},
"p50": {
"mean": 0.07747548896284041,
"min": 0.07182548360524435,
"max": 0.08312549432043648
},
"p90": {
"mean": -0.03741754843230263,
"min": -0.04507556167777899,
"max": -0.029759535186826266
},
"p99": {
"mean": -0.1000082228330611,
"min": -0.10054590154610009,
"max": -0.09947054412002211
}
},
"e2e": {
"mean": {
"mean": 0.02389657805395768,
"min": 0.01846906077214167,
"max": 0.029324095335773688
},
"p50": {
"mean": 0.07396497196591398,
"min": 0.07205509683921338,
"max": 0.07587484709261458
},
"p90": {
"mean": -0.0322944538772641,
"min": -0.037157045612690995,
"max": -0.027431862141837213
},
"p99": {
"mean": -0.047180224796584846,
"min": -0.051216767638022506,
"max": -0.04314368195514718
}
}
},
"legacy_pooled_distribution_bias": {
"ttft": {
"mean": -0.14129871969878843,
"p50": -0.2042045530944649,
"p90": -0.19003454588767263,
"p99": -0.19373351009539902
},
"tpot": {
"mean": 0.03117337438562004,
"p50": 0.07668249597302182,
"p90": -0.03724827658429621,
"p99": -0.10050436157089844
},
"e2e": {
"mean": 0.023867807624916495,
"p50": 0.07536668131278851,
"p90": -0.03100470462321266,
"p99": -0.04909581632382212
}
}
},
"new": {
"trialwise_distribution_bias": [
{
"ttft": {
"mean": -0.07113138559444453,
"p50": -0.1294796632798084,
"p90": -0.11063584772248519,
"p99": -0.08982438463776078
},
"tpot": {
"mean": 0.048585629713612626,
"p50": 0.08234652581183928,
"p90": 0.003101343693636994,
"p99": -0.05952673172238401
},
"e2e": {
"mean": 0.04462683161260099,
"p50": 0.0878889042422101,
"p90": -0.006843953727060337,
"p99": -0.0036641600308731643
}
},
{
"ttft": {
"mean": -0.07051842392629604,
"p50": -0.12771160295843229,
"p90": -0.11011498510890981,
"p99": -0.08861845103384182
},
"tpot": {
"mean": 0.060629878515092876,
"p50": 0.09375745746657616,
"p90": 0.019190079238111446,
"p99": -0.05840233314016777
},
"e2e": {
"mean": 0.05576066061144686,
"p50": 0.09176507061640418,
"p90": 0.003187406757202862,
"p99": 0.004813545234813467
}
}
],
"trialwise_distribution_bias_summary": {
"ttft": {
"mean": {
"mean": -0.07082490476037029,
"min": -0.07113138559444453,
"max": -0.07051842392629604
},
"p50": {
"mean": -0.12859563311912034,
"min": -0.1294796632798084,
"max": -0.12771160295843229
},
"p90": {
"mean": -0.11037541641569751,
"min": -0.11063584772248519,
"max": -0.11011498510890981
},
"p99": {
"mean": -0.0892214178358013,
"min": -0.08982438463776078,
"max": -0.08861845103384182
}
},
"tpot": {
"mean": {
"mean": 0.05460775411435275,
"min": 0.048585629713612626,
"max": 0.060629878515092876
},
"p50": {
"mean": 0.08805199163920771,
"min": 0.08234652581183928,
"max": 0.09375745746657616
},
"p90": {
"mean": 0.01114571146587422,
"min": 0.003101343693636994,
"max": 0.019190079238111446
},
"p99": {
"mean": -0.05896453243127589,
"min": -0.05952673172238401,
"max": -0.05840233314016777
}
},
"e2e": {
"mean": {
"mean": 0.05019374611202393,
"min": 0.04462683161260099,
"max": 0.05576066061144686
},
"p50": {
"mean": 0.08982698742930714,
"min": 0.0878889042422101,
"max": 0.09176507061640418
},
"p90": {
"mean": -0.0018282734849287376,
"min": -0.006843953727060337,
"max": 0.003187406757202862
},
"p99": {
"mean": 0.0005746926019701514,
"min": -0.0036641600308731643,
"max": 0.004813545234813467
}
}
},
"paired_relative_error": [
{
"ttft": {
"mean": 0.3165020807873401,
"p50": 0.008761867404954687,
"p90": 1.027515859973651,
"p99": 4.618320663100516
},
"tpot": {
"mean": 0.09697936002481403,
"p50": 0.08127529560627057,
"p90": 0.21395969552811614,
"p99": 0.6138944058729238
},
"e2e": {
"mean": 0.09434768590630697,
"p50": 0.07464475818134242,
"p90": 0.22980866153598797,
"p99": 0.6060590884554267
}
},
{
"ttft": {
"mean": 0.3247210835745076,
"p50": 0.00808317178222498,
"p90": 1.0875948711956118,
"p99": 4.791245921444296
},
"tpot": {
"mean": 0.10993925696112139,
"p50": 0.09402822239992542,
"p90": 0.22283625990015465,
"p99": 0.6083519770881531
},
"e2e": {
"mean": 0.10658386724241943,
"p50": 0.08490549835786211,
"p90": 0.2445084767702898,
"p99": 0.6267174054324888
}
}
],
"legacy_pooled_distribution_bias": {
"ttft": {
"mean": -0.07082500585057615,
"p50": -0.1284088888361358,
"p90": -0.111136573344055,
"p99": -0.09012556161973535
},
"tpot": {
"mean": 0.05457336599027876,
"p50": 0.0872512146277895,
"p90": 0.01132352325010614,
"p99": -0.05948329733667248
},
"e2e": {
"mean": 0.050164236759908075,
"p50": 0.0912493994395978,
"p90": -0.0004979191794830456,
"p99": -0.0014369075488634014
}
},
"waiting_p99_ms": 1166.1196883368564,
"validity": "GATE_FAIL_DIAGNOSTIC"
}
},
"tp4_rho0p0025": {
"old": {
"trialwise_distribution_bias": [
{
"ttft": {
"mean": 0.025153752214840692,
"p50": -0.04039475356731359,
"p90": -0.05617389473222491,
"p99": -0.07293111999637672
},
"tpot": {
"mean": 0.21469140550335514,
"p50": 0.22290333666146916,
"p90": 0.1721777151300294,
"p99": 0.1622573651896084
},
"e2e": {
"mean": 0.18156895570089593,
"p50": 0.20381782987468136,
"p90": 0.18582117345248075,
"p99": 0.15166386730143064
}
},
{
"ttft": {
"mean": 0.024891430828052088,
"p50": -0.03968933093624011,
"p90": -0.05623634198811063,
"p99": -0.0733553458724018
},
"tpot": {
"mean": 0.21936408831731027,
"p50": 0.2282296491940284,
"p90": 0.17345944256977727,
"p99": 0.16387998107689886
},
"e2e": {
"mean": 0.18499760077470503,
"p50": 0.2077274237083677,
"p90": 0.18741568436689596,
"p99": 0.15614012298902055
}
}
],
"trialwise_distribution_bias_summary": {
"ttft": {
"mean": {
"mean": 0.02502259152144639,
"min": 0.024891430828052088,
"max": 0.025153752214840692
},
"p50": {
"mean": -0.04004204225177685,
"min": -0.04039475356731359,
"max": -0.03968933093624011
},
"p90": {
"mean": -0.05620511836016777,
"min": -0.05623634198811063,
"max": -0.05617389473222491
},
"p99": {
"mean": -0.07314323293438926,
"min": -0.0733553458724018,
"max": -0.07293111999637672
}
},
"tpot": {
"mean": {
"mean": 0.21702774691033272,
"min": 0.21469140550335514,
"max": 0.21936408831731027
},
"p50": {
"mean": 0.22556649292774877,
"min": 0.22290333666146916,
"max": 0.2282296491940284
},
"p90": {
"mean": 0.17281857884990334,
"min": 0.1721777151300294,
"max": 0.17345944256977727
},
"p99": {
"mean": 0.16306867313325363,
"min": 0.1622573651896084,
"max": 0.16387998107689886
}
},
"e2e": {
"mean": {
"mean": 0.1832832782378005,
"min": 0.18156895570089593,
"max": 0.18499760077470503
},
"p50": {
"mean": 0.20577262679152453,
"min": 0.20381782987468136,
"max": 0.2077274237083677
},
"p90": {
"mean": 0.18661842890968836,
"min": 0.18582117345248075,
"max": 0.18741568436689596
},
"p99": {
"mean": 0.1539019951452256,
"min": 0.15166386730143064,
"max": 0.15614012298902055
}
}
},
"legacy_pooled_distribution_bias": {
"ttft": {
"mean": 0.025022574738277282,
"p50": -0.040042171846277425,
"p90": -0.056230097634382616,
"p99": -0.07535478089127973
},
"tpot": {
"mean": 0.2170232618103144,
"p50": 0.22406751004936917,
"p90": 0.1725668492041552,
"p99": 0.1603764334794579
},
"e2e": {
"mean": 0.18328079455378776,
"p50": 0.2057109615696404,
"p90": 0.1869703879211648,
"p99": 0.14838304065885655
}
}
},
"new": {
"trialwise_distribution_bias": [
{
"ttft": {
"mean": 0.07780438024003325,
"p50": 0.028775382574128122,
"p90": 0.025284043303025253,
"p99": 0.02322998991779303
},
"tpot": {
"mean": 0.2271612069590991,
"p50": 0.22290523091840317,
"p90": 0.17920508135226615,
"p99": 0.25623246663646587
},
"e2e": {
"mean": 0.19018228704673856,
"p50": 0.21062858068127338,
"p90": 0.1916437019129637,
"p99": 0.15489591747842305
}
},
{
"ttft": {
"mean": 0.07752858635145746,
"p50": 0.029531653384293997,
"p90": 0.025216206468834785,
"p99": 0.022761760805360378
},
"tpot": {
"mean": 0.2318818586865509,
"p50": 0.22823155170133197,
"p90": 0.1804944929238653,
"p99": 0.2579862802233231
},
"e2e": {
"mean": 0.19363592605420304,
"p50": 0.21456029353388054,
"p90": 0.1932460420729579,
"p99": 0.15938473541043005
}
}
],
"trialwise_distribution_bias_summary": {
"ttft": {
"mean": {
"mean": 0.07766648329574535,
"min": 0.07752858635145746,
"max": 0.07780438024003325
},
"p50": {
"mean": 0.02915351797921106,
"min": 0.028775382574128122,
"max": 0.029531653384293997
},
"p90": {
"mean": 0.02525012488593002,
"min": 0.025216206468834785,
"max": 0.025284043303025253
},
"p99": {
"mean": 0.022995875361576705,
"min": 0.022761760805360378,
"max": 0.02322998991779303
}
},
"tpot": {
"mean": {
"mean": 0.229521532822825,
"min": 0.2271612069590991,
"max": 0.2318818586865509
},
"p50": {
"mean": 0.22556839130986756,
"min": 0.22290523091840317,
"max": 0.22823155170133197
},
"p90": {
"mean": 0.1798497871380657,
"min": 0.17920508135226615,
"max": 0.1804944929238653
},
"p99": {
"mean": 0.2571093734298945,
"min": 0.25623246663646587,
"max": 0.2579862802233231
}
},
"e2e": {
"mean": {
"mean": 0.1919091065504708,
"min": 0.19018228704673856,
"max": 0.19363592605420304
},
"p50": {
"mean": 0.21259443710757697,
"min": 0.21062858068127338,
"max": 0.21456029353388054
},
"p90": {
"mean": 0.1924448719929608,
"min": 0.1916437019129637,
"max": 0.1932460420729579
},
"p99": {
"mean": 0.15714032644442655,
"min": 0.15489591747842305,
"max": 0.15938473541043005
}
}
},
"paired_relative_error": [
{
"ttft": {
"mean": 0.28957904842187726,
"p50": 0.11967766737369243,
"p90": 0.5365251338257485,
"p99": 3.680446479340459
},
"tpot": {
"mean": 0.2360056579049085,
"p50": 0.21333501659187393,
"p90": 0.27188619704554073,
"p99": 0.8481115630045084
},
"e2e": {
"mean": 0.22033571422861917,
"p50": 0.20586761334949796,
"p90": 0.282293891126381,
"p99": 0.6891849392053429
}
},
{
"ttft": {
"mean": 0.28322292774037994,
"p50": 0.12089296869324834,
"p90": 0.5087862765467016,
"p99": 3.551749946511281
},
"tpot": {
"mean": 0.24183792864428846,
"p50": 0.21903748389459432,
"p90": 0.2783964492106047,
"p99": 0.973107877920707
},
"e2e": {
"mean": 0.2239979384097061,
"p50": 0.21064946414808636,
"p90": 0.28840198884356866,
"p99": 0.6983274345617249
}
}
],
"legacy_pooled_distribution_bias": {
"ttft": {
"mean": 0.07766646565061346,
"p50": 0.029153379043297147,
"p90": 0.025222989748299444,
"p99": 0.0205549324689376
},
"tpot": {
"mean": 0.2295170016795841,
"p50": 0.22406940610958842,
"p90": 0.1795965483385546,
"p99": 0.254199450927048
},
"e2e": {
"mean": 0.19190660476104554,
"p50": 0.21253242300694286,
"p90": 0.19279855916666536,
"p99": 0.15160588346499027
}
},
"waiting_p99_ms": 302.3540478261558,
"validity": "PASS_SUBCRITICAL"
}
},
"tp4_rho0p005": {
"old": {
"trialwise_distribution_bias": [
{
"ttft": {
"mean": 0.02817612266853226,
"p50": 0.029949406527602123,
"p90": -0.05512408243570239,
"p99": -0.04532941341343785
},
"tpot": {
"mean": 0.17688090609346543,
"p50": 0.1740795026677315,
"p90": 0.11018831255907795,
"p99": 0.3192029274754835
},
"e2e": {
"mean": 0.1569382150647865,
"p50": 0.1678105991098637,
"p90": 0.12797847572660626,
"p99": 0.16781804606311032
}
},
{
"ttft": {
"mean": 0.02912207227616738,
"p50": 0.027190990310942655,
"p90": -0.056337427761921965,
"p99": -0.04121672646120952
},
"tpot": {
"mean": 0.16462206191460754,
"p50": 0.16195538436894344,
"p90": 0.09988703619210754,
"p99": 0.31352142471539024
},
"e2e": {
"mean": 0.1453260018192159,
"p50": 0.15247501518771847,
"p90": 0.12213267160696055,
"p99": 0.15897391264426655
}
}
],
"trialwise_distribution_bias_summary": {
"ttft": {
"mean": {
"mean": 0.028649097472349817,
"min": 0.02817612266853226,
"max": 0.02912207227616738
},
"p50": {
"mean": 0.02857019841927239,
"min": 0.027190990310942655,
"max": 0.029949406527602123
},
"p90": {
"mean": -0.055730755098812174,
"min": -0.056337427761921965,
"max": -0.05512408243570239
},
"p99": {
"mean": -0.04327306993732369,
"min": -0.04532941341343785,
"max": -0.04121672646120952
}
},
"tpot": {
"mean": {
"mean": 0.1707514840040365,
"min": 0.16462206191460754,
"max": 0.17688090609346543
},
"p50": {
"mean": 0.16801744351833747,
"min": 0.16195538436894344,
"max": 0.1740795026677315
},
"p90": {
"mean": 0.10503767437559275,
"min": 0.09988703619210754,
"max": 0.11018831255907795
},
"p99": {
"mean": 0.3163621760954369,
"min": 0.31352142471539024,
"max": 0.3192029274754835
}
},
"e2e": {
"mean": {
"mean": 0.1511321084420012,
"min": 0.1453260018192159,
"max": 0.1569382150647865
},
"p50": {
"mean": 0.1601428071487911,
"min": 0.15247501518771847,
"max": 0.1678105991098637
},
"p90": {
"mean": 0.1250555736667834,
"min": 0.12213267160696055,
"max": 0.12797847572660626
},
"p99": {
"mean": 0.16339597935368844,
"min": 0.15897391264426655,
"max": 0.16781804606311032
}
}
},
"legacy_pooled_distribution_bias": {
"ttft": {
"mean": 0.028648879997638963,
"p50": 0.028237979190582876,
"p90": -0.055961238578361966,
"p99": -0.045491187615110146
},
"tpot": {
"mean": 0.1707193936623511,
"p50": 0.16851374859025317,
"p90": 0.10492621353626864,
"p99": 0.31231888769471766
},
"e2e": {
"mean": 0.151102823467785,
"p50": 0.1594213531394918,
"p90": 0.1266352410718406,
"p99": 0.15576537699445703
}
}
},
"new": {
"trialwise_distribution_bias": [
{
"ttft": {
"mean": 0.0830717367573255,
"p50": 0.09202782803564245,
"p90": 0.00028495427033840576,
"p99": 0.052715475291315185
},
"tpot": {
"mean": 0.1875681469549417,
"p50": 0.17992592310879849,
"p90": 0.12349339185025797,
"p99": 0.37339354462095886
},
"e2e": {
"mean": 0.16897544744343032,
"p50": 0.18209714770541094,
"p90": 0.13787606868147936,
"p99": 0.18904626199769245
}
},
{
"ttft": {
"mean": 0.08406819180217444,
"p50": 0.08910315304597226,
"p90": -0.0009995435685739617,
"p99": 0.05725053613897321
},
"tpot": {
"mean": 0.1751979803646619,
"p50": 0.1677414318174738,
"p90": 0.11306865958187659,
"p99": 0.3674786553708338
},
"e2e": {
"mean": 0.1572424162428063,
"p50": 0.1665739541099918,
"p90": 0.1319789697978685,
"p99": 0.18004136280321917
}
}
],
"trialwise_distribution_bias_summary": {
"ttft": {
"mean": {
"mean": 0.08356996427974997,
"min": 0.0830717367573255,
"max": 0.08406819180217444
},
"p50": {
"mean": 0.09056549054080736,
"min": 0.08910315304597226,
"max": 0.09202782803564245
},
"p90": {
"mean": -0.00035729464911777795,
"min": -0.0009995435685739617,
"max": 0.00028495427033840576
},
"p99": {
"mean": 0.054983005715144195,
"min": 0.052715475291315185,
"max": 0.05725053613897321
}
},
"tpot": {
"mean": {
"mean": 0.18138306365980178,
"min": 0.1751979803646619,
"max": 0.1875681469549417
},
"p50": {
"mean": 0.17383367746313616,
"min": 0.1677414318174738,
"max": 0.17992592310879849
},
"p90": {
"mean": 0.11828102571606727,
"min": 0.11306865958187659,
"max": 0.12349339185025797
},
"p99": {
"mean": 0.37043609999589633,
"min": 0.3674786553708338,
"max": 0.37339354462095886
}
},
"e2e": {
"mean": {
"mean": 0.1631089318431183,
"min": 0.1572424162428063,
"max": 0.16897544744343032
},
"p50": {
"mean": 0.17433555090770136,
"min": 0.1665739541099918,
"max": 0.18209714770541094
},
"p90": {
"mean": 0.13492751923967394,
"min": 0.1319789697978685,
"max": 0.13787606868147936
},
"p99": {
"mean": 0.1845438124004558,
"min": 0.18004136280321917,
"max": 0.18904626199769245
}
}
},
"paired_relative_error": [
{
"ttft": {
"mean": 0.3546551986110786,
"p50": 0.13992436338556363,
"p90": 0.6114206167819235,
"p99": 3.950640311967927
},
"tpot": {
"mean": 0.20750924221545067,
"p50": 0.19665829732786516,
"p90": 0.3036816181470401,
"p99": 0.6918780517787506
},
"e2e": {
"mean": 0.1987877190488419,
"p50": 0.1867264266897267,
"p90": 0.3097631367891218,
"p99": 0.6627422276019761
}
},
{
"ttft": {
"mean": 0.35485558010616797,
"p50": 0.1421694148145008,
"p90": 0.6220489522097341,
"p99": 3.6653171696017184
},
"tpot": {
"mean": 0.19287611849010033,
"p50": 0.17961331208149112,
"p90": 0.2699038364626021,
"p99": 0.6456553921719683
},
"e2e": {
"mean": 0.18689663758628208,
"p50": 0.17071369418138402,
"p90": 0.2832272665867874,
"p99": 0.6154236743406891
}
}
],
"legacy_pooled_distribution_bias": {
"ttft": {
"mean": 0.08356973519379125,
"p50": 0.09021324737193111,
"p90": -0.0006012940489499138,
"p99": 0.05253708684194205
},
"tpot": {
"mean": 0.1813506819061229,
"p50": 0.1743324539248605,
"p90": 0.11816822907155744,
"p99": 0.3662267201704473
},
"e2e": {
"mean": 0.1630793421767109,
"p50": 0.17360527090575292,
"p90": 0.13652104764058856,
"p99": 0.17677450343779522
}
},
"waiting_p99_ms": 499.9578367015957,
"validity": "PASS_SUBCRITICAL"
}
},
"tp4_rho0p01": {
"old": {
"trialwise_distribution_bias": [
{
"ttft": {
"mean": 0.001069282130605019,
"p50": 0.23624600468603504,
"p90": -0.06243912870195954,
"p99": -0.09510776821394695
},
"tpot": {
"mean": 0.06803177229402764,
"p50": 0.08606085683396018,
"p90": 0.04169520062620168,
"p99": -0.007371970069137671
},
"e2e": {
"mean": 0.07347872011642306,
"p50": 0.12052386744228415,
"p90": 0.03440294047660334,
"p99": 0.038874086941855734
}
},
{
"ttft": {
"mean": -0.003986909712741579,
"p50": 0.22473642702400629,
"p90": -0.06765340173498505,
"p99": -0.09602231432410636
},
"tpot": {
"mean": 0.06548345266052943,
"p50": 0.08325866738979519,
"p90": 0.03667309817451709,
"p99": -0.012893094435078642
},
"e2e": {
"mean": 0.07078275584585807,
"p50": 0.1217320014916679,
"p90": 0.03247676382487049,
"p99": 0.03246095420030834
}
}
],
"trialwise_distribution_bias_summary": {
"ttft": {
"mean": {
"mean": -0.0014588137910682799,
"min": -0.003986909712741579,
"max": 0.001069282130605019
},
"p50": {
"mean": 0.23049121585502066,
"min": 0.22473642702400629,
"max": 0.23624600468603504
},
"p90": {
"mean": -0.06504626521847229,
"min": -0.06765340173498505,
"max": -0.06243912870195954
},
"p99": {
"mean": -0.09556504126902665,
"min": -0.09602231432410636,
"max": -0.09510776821394695
}
},
"tpot": {
"mean": {
"mean": 0.06675761247727854,
"min": 0.06548345266052943,
"max": 0.06803177229402764
},
"p50": {
"mean": 0.08465976211187769,
"min": 0.08325866738979519,
"max": 0.08606085683396018
},
"p90": {
"mean": 0.039184149400359386,
"min": 0.03667309817451709,
"max": 0.04169520062620168
},
"p99": {
"mean": -0.010132532252108157,
"min": -0.012893094435078642,
"max": -0.007371970069137671
}
},
"e2e": {
"mean": {
"mean": 0.07213073798114056,
"min": 0.07078275584585807,
"max": 0.07347872011642306
},
"p50": {
"mean": 0.12112793446697602,
"min": 0.12052386744228415,
"max": 0.1217320014916679
},
"p90": {
"mean": 0.033439852150736915,
"min": 0.03247676382487049,
"max": 0.03440294047660334
},
"p99": {
"mean": 0.03566752057108204,
"min": 0.03246095420030834,
"max": 0.038874086941855734
}
}
},
"legacy_pooled_distribution_bias": {
"ttft": {
"mean": -0.0014652143973501086,
"p50": 0.22936601881498542,
"p90": -0.0646093465409875,
"p99": -0.09621678853313553
},
"tpot": {
"mean": 0.06675609059150077,
"p50": 0.08477302587551214,
"p90": 0.0379183273767328,
"p99": -0.010663954751357074
},
"e2e": {
"mean": 0.07212904317306096,
"p50": 0.12069215463307655,
"p90": 0.03317293927357903,
"p99": 0.0338378271163308
}
}
},
"new": {
"trialwise_distribution_bias": [
{
"ttft": {
"mean": 0.06234024172127873,
"p50": 0.24853946560541346,
"p90": -0.009255094378797821,
"p99": -0.014267710892388628
},
"tpot": {
"mean": 0.10728357759318019,
"p50": 0.09833993314940369,
"p90": 0.08764146258310636,
"p99": 0.07022746965618598
},
"e2e": {
"mean": 0.09915483987796365,
"p50": 0.13866670972485098,
"p90": 0.05932039093418303,
"p99": 0.068986345174611
}
},
{
"ttft": {
"mean": 0.056974583059155086,
"p50": 0.23691543455574957,
"p90": -0.014765152021066699,
"p99": -0.01526395950520138
},
"tpot": {
"mean": 0.10464160330568303,
"p50": 0.09550606196492119,
"p90": 0.08239784923775459,
"p99": 0.06427472725757602
},
"e2e": {
"mean": 0.09639439188714723,
"p50": 0.13989440514740228,
"p90": 0.05734781513815735,
"p99": 0.06238732473827772
}
}
],
"trialwise_distribution_bias_summary": {
"ttft": {
"mean": {
"mean": 0.05965741239021691,
"min": 0.056974583059155086,
"max": 0.06234024172127873
},
"p50": {
"mean": 0.2427274500805815,
"min": 0.23691543455574957,
"max": 0.24853946560541346
},
"p90": {
"mean": -0.01201012319993226,
"min": -0.014765152021066699,
"max": -0.009255094378797821
},
"p99": {
"mean": -0.014765835198795003,
"min": -0.01526395950520138,
"max": -0.014267710892388628
}
},
"tpot": {
"mean": {
"mean": 0.10596259044943161,
"min": 0.10464160330568303,
"max": 0.10728357759318019
},
"p50": {
"mean": 0.09692299755716244,
"min": 0.09550606196492119,
"max": 0.09833993314940369
},
"p90": {
"mean": 0.08501965591043048,
"min": 0.08239784923775459,
"max": 0.08764146258310636
},
"p99": {
"mean": 0.067251098456881,
"min": 0.06427472725757602,
"max": 0.07022746965618598
}
},
"e2e": {
"mean": {
"mean": 0.09777461588255544,
"min": 0.09639439188714723,
"max": 0.09915483987796365
},
"p50": {
"mean": 0.13928055743612663,
"min": 0.13866670972485098,
"max": 0.13989440514740228
},
"p90": {
"mean": 0.05833410303617019,
"min": 0.05734781513815735,
"max": 0.05932039093418303
},
"p99": {
"mean": 0.06568683495644435,
"min": 0.06238732473827772,
"max": 0.068986345174611
}
}
},
"paired_relative_error": [
{
"ttft": {
"mean": 0.31553688775988453,
"p50": 0.11838592821378686,
"p90": 0.6506248643406535,
"p99": 3.727741795520502
},
"tpot": {
"mean": 0.1255843946058368,
"p50": 0.09013696629948227,
"p90": 0.2591848042876693,
"p99": 0.6394926577997524
},
"e2e": {
"mean": 0.12720160501967023,
"p50": 0.09589002059334667,
"p90": 0.26593105160104036,
"p99": 0.6329767056751748
}
},
{
"ttft": {
"mean": 0.3055207717400012,
"p50": 0.11067975650739464,
"p90": 0.6346487333489169,
"p99": 3.723279708208583
},
"tpot": {
"mean": 0.12290433684113768,
"p50": 0.08881300531692908,
"p90": 0.2544711123916822,
"p99": 0.635068592349857
},
"e2e": {
"mean": 0.1238739253923782,
"p50": 0.0938720089907427,
"p90": 0.2617045417392297,
"p99": 0.6208489208233263
}
}
],
"legacy_pooled_distribution_bias": {
"ttft": {
"mean": 0.059650620031540064,
"p50": 0.24159106387124998,
"p90": -0.011548419893895705,
"p99": -0.015475807392170575
},
"tpot": {
"mean": 0.10596101263201793,
"p50": 0.09703754188844527,
"p90": 0.08369800201750718,
"p99": 0.06667813160571406
},
"e2e": {
"mean": 0.09777288053702092,
"p50": 0.1388377217196832,
"p90": 0.058060760567474216,
"p99": 0.06380410696893425
}
},
"waiting_p99_ms": 743.1356935161051,
"validity": "PASS_SUBCRITICAL"
}
}
},
"checks": {
"tp1_ttft_mean_p99_improve_ge_5pp": true,
"tp2_tp4_ttft_e2e_no_abs_regression_gt_5pp": false,
"regressions": [
{
"cell": "tp4_rho0p0025",
"metric": "ttft",
"quantile": "mean",
"old_bias": 0.025022574738277282,
"new_bias": 0.07766646565061346,
"abs_bias_delta_pp": 5.2643890912336175,
"validity": "PASS_SUBCRITICAL"
},
{
"cell": "tp4_rho0p005",
"metric": "ttft",
"quantile": "mean",
"old_bias": 0.028648879997638963,
"new_bias": 0.08356973519379125,
"abs_bias_delta_pp": 5.492085519615229,
"validity": "PASS_SUBCRITICAL"
},
{
"cell": "tp4_rho0p005",
"metric": "ttft",
"quantile": "p50",
"old_bias": 0.028237979190582876,
"new_bias": 0.09021324737193111,
"abs_bias_delta_pp": 6.197526818134823,
"validity": "PASS_SUBCRITICAL"
},
{
"cell": "tp4_rho0p01",
"metric": "ttft",
"quantile": "mean",
"old_bias": -0.0014652143973501086,
"new_bias": 0.059650620031540064,
"abs_bias_delta_pp": 5.8185405634189955,
"validity": "PASS_SUBCRITICAL"
}
],
"trace_gate": false
}
}