kreas
/

MA_benchmarks

Model card Files Files and versions Community

kreas commited on Apr 19

Commit

44d5893

•

1 Parent(s): 1d9b011

Upload /Llama-2-7b-hf-GPTQ-4bit/gptq4bit_batch_size_1_sq_len_256_new_tokens_256/benchmark_report.json with huggingface_hub

Browse files

Files changed (1) hide show

Llama-2-7b-hf-GPTQ-4bit/gptq4bit_batch_size_1_sq_len_256_new_tokens_256/benchmark_report.json +781 -780

Llama-2-7b-hf-GPTQ-4bit/gptq4bit_batch_size_1_sq_len_256_new_tokens_256/benchmark_report.json CHANGED Viewed

@@ -2,284 +2,285 @@
     "prefill": {
         "memory": {
             "unit": "MB",
-            "max_ram": 5657.923584,
             "max_vram": 4980.736,
             "max_reserved": 4494.196736,
             "max_allocated": 4322.724864
         },
         "latency": {
             "unit": "s",
-            "mean": 0.038103712060152346,
-            "stdev": 0.00019952577029451316,
             "values": [
-                0.04095161437988281,
-                0.03870719909667969,
-                0.03872972869873047,
-                0.03864473724365235,
-                0.03871539306640625,
-                0.03868374252319336,
-                0.03822387313842773,
-                0.03794636917114258,
-                0.03805388641357422,
-                0.03798323059082031,
-                0.03814092636108399,
-                0.03803955078125,
-                0.03808665466308594,
-                0.03806719970703125,
-                0.03810508728027344,
-                0.03808563232421875,
-                0.03809075164794922,
-                0.038024192810058595,
-                0.03803647994995117,
-                0.0380313606262207,
-                0.03807436752319336,
-                0.038024192810058595,
-                0.038101249694824216,
-                0.03801497650146484,
-                0.03809692764282226,
-                0.03803750228881836,
-                0.03808358383178711,
-                0.038024192810058595,
-                0.03812249755859375,
-                0.03802316665649414,
-                0.03808988952636719,
-                0.03803238296508789,
-                0.0380682258605957,
-                0.038013950347900394,
-                0.03808563232421875,
-                0.038040576934814455,
-                0.038035457611083984,
-                0.038044830322265626,
-                0.038112255096435545,
-                0.03807027053833008,
-                0.03811532974243164,
-                0.03806412887573242,
-                0.03809075164794922,
-                0.03804876708984375,
-                0.038095073699951174,
-                0.03804159927368164,
-                0.0381214714050293,
-                0.03803852844238281,
-                0.038063102722167966,
-                0.03801190567016602,
-                0.0381317138671875,
-                0.03801804733276367,
-                0.038063102722167966,
-                0.03803750228881836,
-                0.03811328125,
                 0.03800678253173828,
-                0.03805593490600586,
-                0.038043647766113284,
-                0.03803443145751953,
-                0.038029312133789066,
-                0.03808870315551758,
-                0.038059009552001956,
-                0.03811635208129883,
-                0.03802009582519531,
-                0.038095870971679685,
-                0.03803852844238281,
-                0.03808153533935547,
-                0.03806617736816406,
-                0.03806208038330078,
-                0.03803443145751953,
-                0.03808256149291992,
-                0.03804467010498047,
-                0.03811532974243164,
-                0.03804467010498047,
-                0.038118431091308594,
-                0.03803750228881836,
-                0.038095870971679685,
-                0.03803340911865234,
-                0.03808051300048828,
-                0.038013950347900394,
-                0.03810406494140625,
-                0.03802521514892578,
-                0.03803340911865234,
-                0.038056961059570314,
-                0.0380579833984375,
-                0.03803443145751953,
-                0.0381102066040039,
-                0.03807743835449219,
-                0.0380948486328125,
-                0.03805507278442383,
-                0.0381102066040039,
-                0.038002689361572264,
-                0.03808374404907226,
-                0.03803033447265625,
-                0.038056961059570314,
-                0.03802316665649414,
-                0.03805184173583984,
-                0.038100990295410156,
-                0.038114303588867186,
-                0.03803964614868164,
-                0.03810815811157227,
                 0.03801190567016602,
-                0.03809382247924804,
-                0.03801599884033203,
-                0.03806208038330078,
-                0.03806412887573242,
-                0.03812044906616211,
-                0.03803145599365235,
-                0.03811840057373047,
-                0.03804569625854492,
-                0.038084606170654296,
-                0.038046718597412106,
-                0.03805081558227539,
                 0.03800064086914062,
-                0.03808358383178711,
-                0.03807027053833008,
-                0.03803750228881836,
-                0.03811328125,
-                0.03812044906616211,
-                0.03808870315551758,
-                0.03808153533935547,
-                0.03804876708984375,
-                0.03808768081665039,
-                0.03806617736816406,
-                0.03806924819946289,
-                0.03807743835449219,
-                0.03806719970703125,
-                0.038037601470947265,
-                0.03810211181640625,
-                0.03809904098510742,
-                0.03810201644897461,
                 0.03800678253173828,
-                0.03813478469848633,
-                0.03805184173583984,
-                0.038112255096435545,
-                0.038040576934814455,
-                0.03806105422973633,
-                0.038075393676757815,
-                0.038089729309082034,
-                0.03801702499389648,
-                0.03807980728149414,
-                0.03805081558227539,
-                0.03813888168334961,
-                0.03804159927368164,
-                0.0380211181640625,
-                0.03803340911865234,
-                0.03814809417724609,
-                0.03803647994995117,
-                0.038063102722167966,
-                0.03806515121459961,
-                0.038091777801513675,
-                0.038076416015625,
-                0.03805184173583984,
-                0.03806515121459961,
-                0.038071327209472657,
-                0.038160385131835936,
-                0.038133758544921875,
-                0.03806412887573242,
-                0.038114303588867186,
-                0.03801497650146484,
-                0.038109184265136715,
-                0.038076416015625,
-                0.03810406494140625,
-                0.038079486846923825,
-                0.0381399040222168,
-                0.03808256149291992,
-                0.03809894561767578,
-                0.03805081558227539,
-                0.03810201644897461,
-                0.03808256149291992,
-                0.038147071838378906,
-                0.03809075164794922,
-                0.03805286407470703,
-                0.03807129669189453,
-                0.0381399040222168,
-                0.038027263641357424,
-                0.03806412887573242,
-                0.03804467010498047,
-                0.03811942291259766,
-                0.03807846450805664,
-                0.03808870315551758,
-                0.03806003189086914,
-                0.038171646118164065,
-                0.038079521179199216,
-                0.038059200286865234,
-                0.03806105422973633,
-                0.03814912033081055,
-                0.03802521514892578,
-                0.038125568389892575,
-                0.03806105422973633,
-                0.03813683319091797,
-                0.0380313606262207,
-                0.03810508728027344,
-                0.03803955078125,
-                0.03810201644897461,
-                0.03807027053833008,
-                0.03810201644897461,
-                0.03807027053833008,
-                0.038095870971679685,
-                0.03807436752319336,
-                0.03808870315551758,
-                0.03805184173583984,
-                0.03813792037963867,
-                0.03806105422973633,
-                0.03810611343383789,
-                0.038059009552001956,
-                0.03807231903076172,
-                0.038042625427246096,
-                0.03809689712524414,
-                0.0380579833984375,
-                0.0381102066040039,
-                0.03806924819946289,
-                0.03812454223632813,
-                0.03806412887573242,
-                0.03811532974243164,
-                0.038024192810058595,
-                0.03808768081665039,
-                0.038059070587158204,
-                0.03806719970703125,
-                0.03807027053833008,
-                0.03811840057373047,
-                0.03809791946411133,
-                0.03812371063232422,
-                0.03805712127685547,
-                0.03811532974243164,
-                0.038076416015625,
-                0.0381921272277832,
-                0.03813580703735352,
-                0.03812966537475586,
-                0.038101215362548825,
-                0.038130687713623046,
-                0.03811328125,
-                0.038163455963134765,
-                0.038075393676757815,
-                0.03815116882324219,
-                0.0381030387878418,
-                0.03812249755859375,
-                0.03812044906616211,
-                0.03818726348876953,
-                0.038109184265136715,
-                0.03817881774902344,
-                0.038128639221191404,
-                0.03813888168334961,
-                0.03811769485473633,
-                0.03817500686645508,
-                0.03812966537475586,
-                0.038133758544921875,
-                0.03812659072875976,
-                0.03814297485351562,
-                0.0381317138671875,
-                0.0381736946105957,
-                0.038109184265136715,
-                0.038158336639404294,
-                0.038125568389892575,
-                0.03812966537475586,
-                0.038150142669677735,
-                0.038133758544921875,
-                0.03810713577270508,
-                0.03812659072875976,
-                0.03812044906616211,
-                0.03813683319091797,
-                0.03811635208129883,
-                0.03814297485351562
             ]
         },
         "throughput": {
             "unit": "tokens/s",
-            "value": 6718.5055250225
         },
         "energy": null,
         "efficiency": null
@@ -287,23 +288,23 @@
     "decode": {
         "memory": {
             "unit": "MB",
-            "max_ram": 5662.117888,
             "max_vram": 5429.526528,
             "max_reserved": 4942.987264,
             "max_allocated": 4868.495872
         },
         "latency": {
             "unit": "s",
-            "mean": 6.110240773200987,
-            "stdev": 0.007569415092468557,
             "values": [
-                6.102671358108518,
-                6.117810188293455
             ]
         },
         "throughput": {
             "unit": "tokens/s",
-            "value": 41.73321632731872
         },
         "energy": null,
         "efficiency": null
@@ -312,524 +313,524 @@
         "memory": null,
         "latency": {
             "unit": "s",
-            "mean": 0.023961728522356802,
-            "stdev": 0.0001816118850650423,
             "values": [
-                0.0253439998626709,
                 0.024781824111938477,
-                0.02395238494873047,
-                0.024028160095214843,
-                0.02429747200012207,
-                0.024044544219970702,
-                0.023794687271118165,
-                0.023996416091918944,
-                0.024009727478027345,
-                0.024189952850341798,
-                0.023900159835815428,
-                0.023988224029541014,
-                0.023887903213500976,
-                0.023918560028076172,
-                0.023759872436523437,
-                0.023962623596191408,
-                0.023809024810791016,
-                0.02391756820678711,
-                0.024191999435424806,
-                0.023967744827270508,
-                0.023880992889404297,
-                0.023939807891845702,
-                0.02379088020324707,
-                0.0239899845123291,
-                0.02386944007873535,
-                0.023954431533813478,
-                0.023775487899780273,
-                0.024070911407470703,
-                0.0239052791595459,
-                0.023989248275756835,
-                0.0238786563873291,
-                0.02405580711364746,
-                0.02388479995727539,
-                0.02389811134338379,
-                0.023803903579711915,
-                0.023969791412353517,
-                0.023835647583007814,
-                0.024006656646728516,
-                0.023848960876464844,
-                0.024056831359863282,
                 0.02474393653869629,
-                0.024007680892944337,
-                0.023815296173095704,
-                0.02396659278869629,
-                0.023797887802124024,
-                0.0238918399810791,
-                0.02386227226257324,
-                0.02412031936645508,
-                0.023863296508789062,
-                0.023937023162841797,
-                0.023980031967163085,
-                0.024106048583984376,
-                0.023957439422607422,
-                0.024017919540405275,
-                0.023819263458251954,
-                0.02404249572753906,
-                0.023821504592895507,
-                0.02400873565673828,
-                0.023808799743652343,
-                0.02386636734008789,
-                0.023827455520629884,
-                0.02390937614440918,
-                0.023780351638793946,
-                0.023994367599487306,
-                0.023787519454956055,
-                0.02388275146484375,
-                0.023827455520629884,
-                0.023883775711059572,
-                0.02374553680419922,
-                0.02389913558959961,
-                0.023799808502197265,
-                0.024066047668457033,
-                0.023805952072143553,
-                0.023880704879760743,
-                0.023777280807495117,
-                0.02388172721862793,
-                0.023785472869873047,
-                0.023928831100463867,
-                0.023806976318359374,
-                0.02407526397705078,
-                0.023780351638793946,
-                0.0238786563873291,
-                0.023736320495605468,
-                0.023916543960571288,
-                0.0237673282623291,
-                0.023932640075683593,
-                0.023827455520629884,
-                0.024351743698120116,
-                0.02377334403991699,
-                0.023945056915283203,
-                0.023837696075439452,
-                0.02395955276489258,
-                0.023841791152954102,
-                0.023971839904785155,
-                0.02386025619506836,
-                0.02401686477661133,
-                0.023817216873168946,
-                0.02390630340576172,
-                0.0238919677734375,
-                0.024036352157592773,
-                0.02389504051208496,
-                0.024026111602783205,
-                0.023940095901489256,
-                0.02412851142883301,
-                0.02392064094543457,
-                0.02405068778991699,
-                0.023960575103759766,
-                0.024023040771484375,
-                0.024958976745605467,
-                0.024071168899536134,
-                0.023928831100463867,
-                0.024061952590942383,
-                0.02388275146484375,
-                0.024044544219970702,
-                0.02393497657775879,
-                0.024028160095214843,
-                0.023910400390625,
-                0.023990272521972656,
-                0.024379392623901368,
-                0.024094720840454102,
-                0.02388083267211914,
-                0.023864191055297853,
-                0.02368307113647461,
-                0.023803903579711915,
-                0.023766016006469725,
-                0.02387660789489746,
-                0.023656511306762697,
-                0.023888832092285157,
-                0.024129535675048826,
-                0.023797760009765623,
-                0.02373734474182129,
-                0.023809024810791016,
-                0.02371174430847168,
-                0.02384486389160156,
-                0.023744512557983398,
-                0.02388582420349121,
-                0.023737503051757813,
-                0.023852895736694336,
-                0.023778303146362305,
-                0.023842815399169923,
-                0.02368409538269043,
-                0.023802879333496094,
-                0.023684255599975584,
-                0.023837535858154298,
-                0.023759872436523437,
-                0.023819263458251954,
-                0.023743488311767577,
-                0.023801855087280274,
-                0.024103071212768553,
-                0.02383958435058594,
-                0.023816192626953125,
-                0.02390732765197754,
-                0.023888896942138672,
-                0.023957504272460937,
-                0.023989248275756835,
-                0.024377344131469726,
-                0.024015871047973633,
-                0.024034303665161134,
-                0.023910400390625,
-                0.02390732765197754,
-                0.023974912643432617,
-                0.024057855606079103,
-                0.024003583908081053,
-                0.023962623596191408,
-                0.023933952331542968,
-                0.02408243179321289,
-                0.023882879257202148,
-                0.02399731254577637,
-                0.0239052791595459,
-                0.023980031967163085,
-                0.023875583648681642,
-                0.023987199783325194,
-                0.024209407806396483,
-                0.02392166328430176,
-                0.02389401626586914,
-                0.023988224029541014,
-                0.02394419288635254,
-                0.023998464584350586,
-                0.024014848709106446,
-                0.024066047668457033,
-                0.023940095901489256,
-                0.02391449546813965,
-                0.023852031707763673,
-                0.02387763214111328,
-                0.023799808502197265,
-                0.023855104446411132,
-                0.023832672119140624,
-                0.023948192596435547,
-                0.02369945526123047,
-                0.023826431274414063,
-                0.02371993637084961,
-                0.0238786563873291,
-                0.023780351638793946,
-                0.02384998321533203,
-                0.023835647583007814,
-                0.02393622398376465,
-                0.02388150405883789,
-                0.0238919677734375,
-                0.023831775665283203,
-                0.02389072036743164,
-                0.024645631790161132,
-                0.023883775711059572,
-                0.02391244888305664,
-                0.024015871047973633,
-                0.023908351898193358,
-                0.023846912384033202,
-                0.023756032943725587,
-                0.023866111755371094,
-                0.02384604835510254,
-                0.023887712478637694,
-                0.023863296508789062,
-                0.02387046432495117,
-                0.023871488571166992,
-                0.023964927673339843,
-                0.023856895446777344,
-                0.023814144134521483,
-                0.02388172721862793,
-                0.02392268753051758,
-                0.02385817527770996,
-                0.02392678451538086,
-                0.02387353515625,
-                0.02388582420349121,
-                0.023864320755004883,
-                0.023829504013061522,
-                0.023843839645385743,
-                0.02385305595397949,
-                0.023846912384033202,
-                0.02389401626586914,
-                0.023847936630249023,
-                0.02393907165527344,
-                0.023800832748413086,
-                0.024054784774780274,
-                0.024442880630493165,
-                0.023962623596191408,
-                0.023987199783325194,
-                0.024026111602783205,
-                0.023985183715820313,
-                0.02401686477661133,
-                0.023964672088623046,
-                0.023983104705810547,
-                0.02394726371765137,
-                0.023989248275756835,
-                0.02390937614440918,
-                0.02385327911376953,
-                0.023824159622192382,
-                0.023871488571166992,
-                0.023848960876464844,
-                0.02389094352722168,
-                0.0238786563873291,
-                0.02389423942565918,
-                0.023853855133056642,
-                0.0239237117767334,
-                0.023835647583007814,
-                0.02390118408203125,
-                0.023842815399169923,
-                0.023966720581054687,
-                0.024193023681640623,
-                0.023974912643432617,
-                0.024176864624023436,
-                0.023990047454833983,
-                0.02409779167175293,
-                0.023973888397216796,
-                0.024146944046020507,
-                0.02393600082397461,
-                0.023981056213378905,
-                0.02389606475830078,
-                0.023997440338134765,
-                0.023938112258911133,
-                0.02405881690979004,
-                0.023966720581054687,
-                0.02415001678466797,
-                0.023940095901489256,
-                0.023975936889648438,
-                0.0238786563873291,
-                0.023950336456298828,
-                0.023908351898193358,
-                0.024138751983642577,
-                0.023948287963867186,
-                0.024159231185913087,
-                0.02393600082397461,
-                0.024027135848999022,
-                0.02389401626586914,
-                0.023983104705810547,
-                0.023998464584350586,
-                0.02408448028564453,
-                0.023976959228515626,
-                0.02414080047607422,
-                0.023940095901489256,
-                0.024026111602783205,
-                0.0238786563873291,
-                0.024060928344726562,
-                0.02389401626586914,
-                0.023965696334838867,
-                0.02390937614440918,
-                0.024193023681640623,
-                0.023964672088623046,
-                0.023989248275756835,
-                0.023908351898193358,
-                0.023996416091918944,
-                0.023932928085327147,
-                0.023994367599487306,
-                0.023970815658569337,
-                0.024164352416992187,
-                0.02389401626586914,
-                0.023958528518676758,
-                0.023868415832519533,
-                0.024016895294189454,
-                0.02390220832824707,
-                0.023992319107055664,
-                0.023953407287597657,
-                0.02408448028564453,
-                0.02390323257446289,
-                0.024094720840454102,
-                0.023946239471435548,
-                0.024004608154296874,
-                0.023969791412353517,
-                0.024028160095214843,
-                0.024040447235107423,
-                0.024211456298828125,
-                0.024032255172729493,
-                0.02408038330078125,
-                0.02390323257446289,
-                0.023969791412353517,
-                0.02390220832824707,
-                0.024040447235107423,
-                0.023839744567871093,
-                0.02406809616088867,
-                0.023964672088623046,
-                0.02405171203613281,
-                0.02391551971435547,
-                0.023994367599487306,
-                0.023953407287597657,
-                0.024029184341430664,
-                0.023968767166137696,
-                0.024359935760498046,
-                0.024034303665161134,
-                0.024164352416992187,
-                0.02404249572753906,
-                0.02407731246948242,
-                0.024031232833862305,
-                0.024139776229858398,
-                0.023994367599487306,
-                0.02428620719909668,
-                0.02405068778991699,
-                0.02409164810180664,
-                0.024049760818481446,
-                0.02420623970031738,
-                0.024019968032836913,
-                0.0241582088470459,
-                0.024014848709106446,
-                0.024191999435424806,
-                0.024088672637939453,
-                0.02402499198913574,
-                0.023993343353271485,
-                0.0241213436126709,
-                0.02394316864013672,
-                0.024005632400512695,
-                0.023971839904785155,
-                0.024213504791259766,
-                0.02415939140319824,
-                0.024047456741333007,
-                0.023986175537109376,
-                0.02408755111694336,
-                0.02617241668701172,
-                0.023987199783325194,
-                0.02391244888305664,
-                0.02412031936645508,
-                0.024035327911376952,
-                0.024019968032836913,
-                0.024018943786621092,
-                0.024071168899536134,
-                0.02395356750488281,
-                0.024024927139282226,
-                0.023969791412353517,
-                0.024028160095214843,
-                0.023969791412353517,
-                0.0240600643157959,
-                0.02396860885620117,
-                0.023976959228515626,
-                0.023925760269165038,
-                0.024028160095214843,
-                0.02390425682067871,
-                0.023961599349975587,
-                0.02434662437438965,
-                0.023997440338134765,
-                0.023872512817382813,
-                0.023988224029541014,
-                0.023925760269165038,
-                0.023957504272460937,
-                0.023958784103393554,
-                0.023963392257690428,
-                0.0238919677734375,
-                0.02408243179321289,
-                0.023950336456298828,
-                0.023938047409057618,
-                0.023836671829223634,
-                0.023990272521972656,
-                0.023888896942138672,
-                0.02395238494873047,
-                0.023810047149658203,
-                0.02389401626586914,
-                0.02388787269592285,
-                0.0239554557800293,
-                0.02437945556640625,
-                0.02401273536682129,
-                0.023856128692626953,
-                0.02400057601928711,
-                0.02389504051208496,
-                0.024059839248657226,
-                0.023954431533813478,
-                0.023964672088623046,
-                0.023872512817382813,
-                0.023946239471435548,
-                0.023863296508789062,
-                0.024000511169433594,
-                0.02391449546813965,
-                0.023977983474731446,
-                0.02393497657775879,
-                0.024038400650024414,
-                0.023812095642089845,
-                0.02392166328430176,
-                0.02384089660644531,
-                0.023884672164916992,
-                0.02387353515625,
-                0.023984128952026368,
-                0.023942144393920898,
-                0.023980031967163085,
-                0.024396799087524415,
-                0.024057855606079103,
-                0.02403436851501465,
-                0.02400454330444336,
-                0.02393907165527344,
-                0.023990272521972656,
-                0.023937215805053712,
-                0.02407302474975586,
-                0.023938047409057618,
-                0.024013824462890625,
-                0.023929887771606446,
-                0.02401686477661133,
-                0.02388172721862793,
-                0.02391756820678711,
-                0.02388479995727539,
-                0.024062976837158204,
-                0.0239554557800293,
-                0.024070144653320313,
-                0.023916543960571288,
-                0.024040447235107423,
-                0.02386944007873535,
-                0.02395136070251465,
-                0.023856128692626953,
-                0.024017919540405275,
-                0.023957504272460937,
-                0.023938047409057618,
-                0.023813119888305666,
-                0.023937023162841797,
-                0.024341503143310548,
-                0.02395136070251465,
-                0.02386124801635742,
-                0.023949312210083007,
-                0.024015871047973633,
-                0.024032255172729493,
-                0.02393606376647949,
-                0.023963584899902343,
-                0.023837696075439452,
-                0.02391756820678711,
-                0.023832576751708984,
-                0.02389401626586914,
-                0.024013824462890625,
-                0.02395238494873047,
-                0.02394726371765137,
-                0.023880704879760743,
-                0.02388796806335449,
-                0.023943071365356446,
-                0.02385305595397949,
-                0.02391756820678711,
-                0.023820287704467775,
-                0.024003583908081053,
-                0.023871616363525392,
-                0.02395737648010254,
-                0.02389094352722168,
-                0.02391449546813965,
-                0.024416255950927734,
-                0.02390630340576172,
-                0.02389311981201172,
-                0.02414681625366211,
-                0.024004608154296874,
-                0.024001535415649415,
-                0.024052736282348632,
-                0.024069120407104492,
-                0.02390323257446289,
-                0.024038400650024414,
-                0.02385817527770996,
-                0.02396067237854004,
-                0.023853984832763672,
-                0.02394726371765137,
-                0.023809024810791016,
-                0.02389299201965332,
-                0.023781375885009767,
-                0.023815168380737304,
-                0.023869600296020508,
-                0.02380988883972168,
-                0.02385408020019531,
-                0.023900159835815428,
-                0.02379267120361328,
-                0.023824352264404297,
-                0.02387455940246582,
-                0.023883775711059572,
-                0.023756799697875978,
-                0.023805952072143553,
-                0.02387660789489746
             ]
         },
         "throughput": {
             "unit": "tokens/s",
-            "value": 41.733216327318736
         },
         "energy": null,
         "efficiency": null

     "prefill": {
         "memory": {
             "unit": "MB",
+            "max_ram": 5730.021376,
             "max_vram": 4980.736,
             "max_reserved": 4494.196736,
             "max_allocated": 4322.724864
         },
         "latency": {
             "unit": "s",
+            "mean": 0.037973927353367694,
+            "stdev": 0.00020372673898222884,
             "values": [
+                0.04081110382080078,
+                0.03922431945800781,
+                0.038899711608886715,
+                0.037784576416015625,
+                0.037972991943359374,
+                0.03784601593017578,
+                0.037940223693847655,
+                0.03788828659057617,
+                0.03794124984741211,
+                0.03789219284057617,
+                0.03799039840698242,
+                0.03791155242919922,
+                0.03793305587768555,
+                0.037935104370117184,
+                0.037978111267089845,
+                0.037928958892822266,
+                0.03794739151000977,
+                0.0379607048034668,
+                0.03794432067871094,
+                0.03791974258422852,
+                0.03798015975952149,
+                0.03787366485595703,
+                0.037961727142333986,
+                0.03789945602416992,
+                0.037959678649902344,
+                0.037923839569091795,
+                0.03815628814697265,
+                0.03793203353881836,
+                0.03791974258422852,
+                0.037920768737792966,
+                0.03794739151000977,
+                0.037928958892822266,
+                0.037959678649902344,
+                0.037940223693847655,
+                0.03795455932617187,
+                0.037940223693847655,
+                0.03798220825195313,
+                0.037926910400390625,
                 0.03800678253173828,
+                0.03791155242919922,
+                0.03796815872192383,
+                0.03792201614379883,
+                0.037928958892822266,
+                0.037926910400390625,
+                0.037958656311035156,
+                0.03794124984741211,
                 0.03801190567016602,
+                0.03788390350341797,
+                0.037966846466064456,
+                0.03793407821655274,
+                0.03798220825195313,
+                0.037969921112060545,
+                0.037978111267089845,
+                0.0379279670715332,
+                0.037937152862548826,
+                0.03790848159790039,
+                0.037958656311035156,
+                0.03790848159790039,
+                0.03798732757568359,
+                0.03792911911010742,
+                0.037976062774658204,
+                0.03798015975952149,
+                0.03792998504638672,
+                0.03789619064331055,
+                0.03798220825195313,
+                0.037937152862548826,
+                0.037959678649902344,
+                0.03791667175292969,
+                0.03798425674438476,
+                0.03791360092163086,
+                0.03799347305297852,
+                0.03794636917114258,
+                0.037988353729248046,
+                0.03794841766357422,
+                0.037943294525146484,
+                0.0379156494140625,
+                0.03796889495849609,
+                0.03792998504638672,
+                0.03796377563476563,
+                0.03793407821655274,
+                0.0379576301574707,
+                0.037943294525146484,
+                0.03799961471557617,
+                0.037958656311035156,
+                0.037997566223144534,
+                0.03789823913574219,
+                0.03799859237670898,
+                0.037937152862548826,
+                0.03798732757568359,
+                0.037904384613037106,
+                0.0379576301574707,
+                0.03793203353881836,
+                0.038002689361572264,
+                0.03794841766357422,
+                0.03799552154541016,
+                0.03792281723022461,
+                0.03798015975952149,
+                0.037901313781738284,
+                0.0379422721862793,
+                0.03793100738525391,
+                0.03798220825195313,
+                0.03792281723022461,
+                0.037956607818603515,
+                0.037904384613037106,
+                0.037953536987304685,
+                0.03793920135498047,
+                0.03796275329589844,
+                0.037926910400390625,
+                0.03798732757568359,
+                0.037933345794677734,
+                0.037986305236816405,
+                0.037940223693847655,
+                0.037978111267089845,
+                0.037917697906494144,
+                0.03798220825195313,
+                0.03791872024536133,
+                0.0379504623413086,
+                0.03791360092163086,
+                0.0379791374206543,
+                0.0379156494140625,
+                0.03799961471557617,
+                0.037904384613037106,
+                0.038018272399902346,
+                0.03793209457397461,
+                0.037988353729248046,
+                0.037905406951904294,
+                0.03794841766357422,
+                0.03790848159790039,
+                0.03797615814208984,
+                0.03793817520141601,
+                0.03797094345092773,
+                0.03789209747314453,
+                0.03799347305297852,
+                0.03795455932617187,
+                0.03794956970214844,
+                0.03792998504638672,
+                0.038004737854003906,
+                0.03791257476806641,
+                0.037978111267089845,
+                0.03796480178833008,
+                0.038004737854003906,
+                0.03794841766357422,
+                0.037959678649902344,
+                0.03791158294677734,
+                0.037958751678466796,
+                0.037935104370117184,
+                0.037953536987304685,
+                0.03793920135498047,
+                0.037989376068115234,
+                0.0379576301574707,
+                0.037976062774658204,
+                0.03793920135498047,
+                0.037956607818603515,
+                0.037953536987304685,
+                0.037978111267089845,
+                0.03794745635986328,
+                0.03799478530883789,
+                0.037940223693847655,
+                0.03798527908325195,
+                0.03793407821655274,
+                0.03798425674438476,
+                0.03794432067871094,
+                0.03796275329589844,
+                0.03791974258422852,
+                0.03796377563476563,
+                0.037917697906494144,
+                0.037996543884277346,
+                0.037961727142333986,
+                0.03799961471557617,
+                0.03794944000244141,
+                0.03797708892822266,
+                0.037937152862548826,
+                0.037975040435791016,
+                0.03793203353881836,
+                0.03796582412719727,
+                0.037921825408935544,
+                0.037997566223144534,
+                0.037940479278564455,
+                0.03797708892822266,
+                0.03795558547973633,
+                0.037961727142333986,
+                0.037943294525146484,
+                0.037975040435791016,
+                0.03789209747314453,
+                0.03796275329589844,
+                0.03795276641845703,
+                0.03802624130249024,
+                0.03790335845947266,
+                0.03792486572265625,
+                0.037951488494873044,
+                0.03801001739501953,
+                0.037935104370117184,
+                0.03800678253173828,
+                0.03789823913574219,
+                0.0379576301574707,
+                0.03793305587768555,
+                0.03800476837158203,
+                0.03793817520141601,
                 0.03800064086914062,
+                0.03794944000244141,
+                0.03797216033935547,
+                0.037956607818603515,
+                0.03799039840698242,
+                0.037927169799804684,
+                0.038019073486328124,
+                0.0379607048034668,
+                0.03799961471557617,
+                0.03793407821655274,
+                0.037994495391845705,
+                0.03797196960449219,
+                0.03797196960449219,
+                0.03794345474243164,
+                0.03799552154541016,
+                0.037978111267089845,
+                0.038004737854003906,
+                0.038129856109619144,
+                0.038046718597412106,
+                0.037953536987304685,
+                0.03794841766357422,
+                0.03791667175292969,
+                0.03799039840698242,
+                0.03792588806152344,
+                0.03802214431762695,
+                0.03787673568725586,
+                0.03797401428222656,
+                0.03795254516601562,
+                0.03795251083374023,
+                0.037921791076660154,
+                0.03798015975952149,
+                0.03794432067871094,
+                0.03794841766357422,
+                0.037930049896240235,
+                0.03800166320800781,
+                0.03794944000244141,
                 0.03800678253173828,
+                0.03790643310546875,
+                0.03801190567016602,
+                0.037923839569091795,
+                0.038005760192871094,
+                0.03795455932617187,
+                0.03798220825195313,
+                0.037940223693847655,
+                0.037994529724121096,
+                0.037901313781738284,
+                0.03802214431762695,
+                0.037923969268798825,
+                0.03798732757568359,
+                0.037905406951904294,
+                0.03796275329589844,
+                0.03798425674438476,
+                0.03799347305297852,
+                0.03793638229370117,
+                0.03801599884033203,
+                0.03795558547973633,
+                0.03798732757568359,
+                0.037928958892822266,
+                0.03797196960449219,
+                0.03794944000244141,
+                0.03801804733276367,
+                0.0379422721862793,
+                0.038013950347900394,
+                0.037951519012451175,
+                0.03792588806152344,
+                0.03792486572265625
             ]
         },
         "throughput": {
             "unit": "tokens/s",
+            "value": 6741.467576365835
         },
         "energy": null,
         "efficiency": null
     "decode": {
         "memory": {
             "unit": "MB",
+            "max_ram": 5732.118528,
             "max_vram": 5429.526528,
             "max_reserved": 4942.987264,
             "max_allocated": 4868.495872
         },
         "latency": {
             "unit": "s",
+            "mean": 6.314276854515072,
+            "stdev": 0.019472382545473454,
             "values": [
+                6.294804471969598,
+                6.333749237060545
             ]
         },
         "throughput": {
             "unit": "tokens/s",
+            "value": 40.38467204960459
         },
         "energy": null,
         "efficiency": null
         "memory": null,
         "latency": {
             "unit": "s",
+            "mean": 0.024761870017706165,
+            "stdev": 0.00019105811774383218,
             "values": [
+                0.026012672424316406,
+                0.02526310348510742,
+                0.02508595275878906,
+                0.024929279327392577,
+                0.024740863800048828,
+                0.025167871475219726,
+                0.024567808151245117,
+                0.024818687438964843,
+                0.024707071304321288,
+                0.02472038459777832,
+                0.024630271911621093,
+                0.02468659210205078,
+                0.02455449676513672,
+                0.02469494438171387,
+                0.02455228805541992,
+                0.024799232482910157,
+                0.024631296157836914,
+                0.02467840003967285,
+                0.024525823593139647,
+                0.02469990348815918,
+                0.024458240509033204,
+                0.02456985664367676,
+                0.024557567596435546,
+                0.024769535064697267,
+                0.02448896026611328,
+                0.024729600906372072,
+                0.024517631530761717,
+                0.025474048614501952,
+                0.024614912033081054,
+                0.024964096069335938,
+                0.02454732894897461,
+                0.02475132751464844,
+                0.024513311386108398,
+                0.024649728775024415,
+                0.024556543350219725,
+                0.02468351936340332,
+                0.024663040161132813,
+                0.024620128631591798,
+                0.02449705505371094,
+                0.024846336364746095,
+                0.024626176834106447,
+                0.024715263366699217,
+                0.024599552154541016,
+                0.024604671478271483,
+                0.024534015655517577,
+                0.024649728775024415,
+                0.024558591842651366,
+                0.024730623245239256,
+                0.024590335845947265,
+                0.024835071563720702,
+                0.024581119537353514,
+                0.024627199172973634,
+                0.02454425621032715,
+                0.024709280014038087,
+                0.024550239562988282,
+                0.02476857566833496,
+                0.024537023544311524,
+                0.024766464233398438,
+                0.024571903228759767,
+                0.02466099166870117,
+                0.02447871971130371,
+                0.02469273567199707,
+                0.02452992057800293,
+                0.024738815307617186,
+                0.024506368637084962,
+                0.025068544387817384,
+                0.024623104095458984,
+                0.024592384338378907,
+                0.02452070426940918,
+                0.02467430305480957,
+                0.024605695724487304,
+                0.02472550392150879,
+                0.024604671478271483,
+                0.024814592361450196,
+                0.024645631790161132,
+                0.02469068717956543,
+                0.024625152587890626,
+                0.024662015914916992,
+                0.024560640335083008,
+                0.024707071304321288,
+                0.024622079849243163,
+                0.02477670478820801,
+                0.024608768463134766,
+                0.024658943176269533,
+                0.02455449676513672,
+                0.0247142391204834,
+                0.0245483512878418,
+                0.024799232482910157,
+                0.024650751113891603,
+                0.024799232482910157,
+                0.024658943176269533,
+                0.024729600906372072,
+                0.024624128341674805,
+                0.024707071304321288,
+                0.024650751113891603,
+                0.02489241600036621,
+                0.02465279960632324,
+                0.024761344909667967,
+                0.02467635154724121,
+                0.02469273567199707,
+                0.024613887786865234,
+                0.024715263366699217,
+                0.024653823852539062,
+                0.024830976486206056,
+                0.02457804870605469,
+                0.02474496078491211,
+                0.024624128341674805,
+                0.024663040161132813,
+                0.02628812789916992,
+                0.024729600906372072,
+                0.02451456069946289,
+                0.0251013126373291,
+                0.024565759658813476,
+                0.024689664840698244,
+                0.024602624893188478,
+                0.024763391494750975,
+                0.024607744216918945,
+                0.024680448532104493,
+                0.024555519104003908,
+                0.02497331237792969,
+                0.024580095291137697,
+                0.024632320404052735,
+                0.024567808151245117,
+                0.024649728775024415,
+                0.02455449676513672,
+                0.024623104095458984,
+                0.02451046371459961,
+                0.024556543350219725,
+                0.02487398338317871,
+                0.024567808151245117,
+                0.02467532730102539,
+                0.02472755241394043,
+                0.024605695724487304,
+                0.024679424285888672,
+                0.024644800186157226,
+                0.024553279876708984,
+                0.024535039901733398,
+                0.02464460754394531,
+                0.024654848098754883,
+                0.024633344650268556,
+                0.02453196716308594,
+                0.024617984771728517,
+                0.02451968002319336,
+                0.024526847839355468,
+                0.024532991409301756,
+                0.024698879241943358,
+                0.024689664840698244,
+                0.024653823852539062,
+                0.024912895202636717,
+                0.024601600646972657,
+                0.02456166458129883,
+                0.024633344650268556,
+                0.024517631530761717,
+                0.024600576400756836,
+                0.024667135238647463,
+                0.02467020797729492,
+                0.024633344650268556,
+                0.02462822341918945,
+                0.024580095291137697,
+                0.024630271911621093,
+                0.024606719970703125,
+                0.02473676872253418,
+                0.02470809555053711,
+                0.024758272171020508,
+                0.02464358329772949,
+                0.02469273567199707,
+                0.024653823852539062,
+                0.024707071304321288,
+                0.024603647232055666,
+                0.02471833610534668,
+                0.02473472023010254,
+                0.02513715171813965,
+                0.024994815826416016,
+                0.02467430305480957,
+                0.024594432830810548,
+                0.02471731185913086,
+                0.024566783905029296,
+                0.024748031616210937,
+                0.02469171142578125,
+                0.02468556785583496,
+                0.024613887786865234,
+                0.024635391235351564,
+                0.024567808151245117,
+                0.02462822341918945,
+                0.024608768463134766,
+                0.024707071304321288,
+                0.02467737579345703,
+                0.024715263366699217,
+                0.02469990348815918,
+                0.02469068717956543,
+                0.024599552154541016,
+                0.024700927734375,
+                0.024603647232055666,
+                0.024641536712646486,
+                0.024603647232055666,
+                0.024771583557128905,
+                0.024629247665405272,
+                0.02471014404296875,
+                0.024673280715942384,
+                0.024664064407348633,
+                0.02512281608581543,
+                0.02512281608581543,
+                0.02472550392150879,
+                0.02483404731750488,
+                0.02475315284729004,
+                0.02469273567199707,
+                0.024663040161132813,
+                0.02469171142578125,
+                0.02456166458129883,
+                0.024665088653564454,
+                0.02467433547973633,
+                0.0247490234375,
+                0.024640512466430665,
+                0.024672256469726563,
+                0.02465996742248535,
+                0.02468659210205078,
+                0.024594432830810548,
+                0.024654848098754883,
+                0.02465996742248535,
+                0.02470911979675293,
+                0.02467737579345703,
+                0.024697856903076174,
+                0.02467840003967285,
+                0.02466815948486328,
+                0.024556543350219725,
+                0.024680448532104493,
+                0.024577024459838868,
+                0.02471833610534668,
+                0.024654848098754883,
+                0.024616960525512696,
+                0.024597503662109374,
+                0.02465996742248535,
+                0.025066495895385742,
+                0.024624128341674805,
+                0.024649728775024415,
+                0.02472652816772461,
+                0.024632320404052735,
+                0.024634368896484377,
+                0.024620031356811522,
+                0.02458624076843262,
+                0.02456166458129883,
+                0.024656896591186524,
+                0.024757247924804687,
+                0.024804351806640625,
+                0.024642559051513673,
+                0.024599552154541016,
+                0.024559616088867187,
+                0.024626176834106447,
+                0.024521728515625,
+                0.024617984771728517,
+                0.024571903228759767,
+                0.024695808410644532,
+                0.02468454360961914,
+                0.02466099166870117,
+                0.0245483512878418,
+                0.024750080108642578,
+                0.024806400299072266,
+                0.024817663192749022,
+                0.02511564826965332,
+                0.02488217544555664,
+                0.02490060806274414,
+                0.024846336364746095,
+                0.025012224197387696,
                 0.024781824111938477,
+                0.024856576919555663,
+                0.024821760177612305,
+                0.025061376571655275,
+                0.024885248184204102,
+                0.024962047576904296,
+                0.024800256729125978,
+                0.025069568634033205,
+                0.024823808670043947,
+                0.0249354248046875,
+                0.02489139175415039,
+                0.025007104873657225,
+                0.024864767074584963,
+                0.024993791580200195,
+                0.024839168548583986,
+                0.025040895462036132,
+                0.024802303314208983,
+                0.024955904006958008,
+                0.024859647750854492,
+                0.025076736450195314,
+                0.02490060806274414,
+                0.024975360870361327,
+                0.0248985595703125,
+                0.024964096069335938,
+                0.024828927993774414,
+                0.024929279327392577,
+                0.0248985595703125,
+                0.02510857582092285,
+                0.024991647720336914,
+                0.024991743087768553,
+                0.02490777587890625,
+                0.025100288391113282,
+                0.02490572738647461,
+                0.02494156837463379,
+                0.02484121513366699,
+                0.024995840072631836,
+                0.024819711685180663,
+                0.024875232696533203,
+                0.02490652847290039,
+                0.024971263885498047,
+                0.024922111511230468,
+                0.024871936798095705,
+                0.02482585525512695,
+                0.025082880020141602,
+                0.024870912551879884,
+                0.02485862350463867,
+                0.024680448532104493,
+                0.024856576919555663,
+                0.024679424285888672,
+                0.02470195198059082,
+                0.02464358329772949,
+                0.024786943435668944,
+                0.02467840003967285,
+                0.024836095809936523,
+                0.02473472023010254,
+                0.024855712890625,
+                0.02461782455444336,
+                0.02467635154724121,
+                0.02472047996520996,
+                0.024917919158935545,
+                0.024814592361450196,
+                0.024870912551879884,
+                0.024859647750854492,
+                0.025007104873657225,
+                0.024844287872314453,
+                0.024896511077880858,
+                0.024838144302368165,
+                0.02503481674194336,
+                0.02486675262451172,
+                0.024863744735717775,
+                0.024863744735717775,
+                0.02502348709106445,
+                0.024959999084472655,
+                0.02490675163269043,
+                0.024852479934692383,
+                0.024988672256469727,
+                0.024844287872314453,
+                0.024870912551879884,
+                0.024819711685180663,
+                0.024954879760742187,
+                0.02472038459777832,
+                0.024748031616210937,
+                0.02467020797729492,
+                0.024798208236694336,
+                0.02476972770690918,
+                0.02478060722351074,
+                0.024688640594482423,
+                0.024930303573608398,
+                0.02472550392150879,
+                0.02467532730102539,
+                0.0247459831237793,
+                0.02489241600036621,
+                0.024819711685180663,
+                0.02482601547241211,
+                0.02474275207519531,
+                0.02493440055847168,
+                0.024655872344970704,
+                0.024773632049560547,
+                0.02472857666015625,
+                0.0248668155670166,
+                0.0265850887298584,
+                0.024808448791503908,
+                0.024797183990478516,
+                0.024992767333984374,
+                0.024896511077880858,
+                0.024829952239990235,
                 0.02474393653869629,
+                0.024976383209228514,
+                0.024819711685180663,
+                0.02490572738647461,
+                0.024838144302368165,
+                0.0248668155670166,
+                0.024803327560424804,
+                0.024790016174316407,
+                0.024782848358154298,
+                0.024819711685180663,
+                0.024853504180908204,
+                0.024810495376586913,
+                0.024773632049560547,
+                0.024797183990478516,
+                0.02509721565246582,
+                0.02471116828918457,
+                0.024664064407348633,
+                0.02487708854675293,
+                0.024827871322631836,
+                0.024830976486206056,
+                0.02465996742248535,
+                0.024821760177612305,
+                0.024786943435668944,
+                0.024799232482910157,
+                0.024667135238647463,
+                0.024790016174316407,
+                0.024829952239990235,
+                0.024814592361450196,
+                0.024697856903076174,
+                0.024811519622802734,
+                0.02470195198059082,
+                0.024791040420532227,
+                0.024827903747558593,
+                0.02485043144226074,
+                0.025235456466674806,
+                0.024893503189086914,
+                0.024790975570678712,
+                0.024878080368041993,
+                0.024763391494750975,
+                0.024835071563720702,
+                0.02476851272583008,
+                0.02487295913696289,
+                0.024861696243286133,
+                0.024864767074584963,
+                0.02468351936340332,
+                0.024808448791503908,
+                0.024796159744262695,
+                0.02491494369506836,
+                0.024787967681884765,
+                0.024795135498046874,
+                0.024819711685180663,
+                0.024855552673339845,
+                0.02478489685058594,
+                0.024843263626098632,
+                0.02469990348815918,
+                0.024803327560424804,
+                0.024704000473022462,
+                0.02484121513366699,
+                0.02528563117980957,
+                0.024919040679931642,
+                0.02474291229248047,
+                0.02489958381652832,
+                0.024863744735717775,
+                0.025168895721435547,
+                0.024731647491455077,
+                0.024769535064697267,
+                0.024862720489501954,
+                0.02490163230895996,
+                0.024761344909667967,
+                0.024807424545288087,
+                0.02473369598388672,
+                0.024770559310913084,
+                0.024756223678588866,
+                0.024944639205932616,
+                0.024943616867065428,
+                0.02488934326171875,
+                0.024731775283813477,
+                0.02476326370239258,
+                0.024723455429077147,
+                0.024748031616210937,
+                0.02463641548156738,
+                0.024769535064697267,
+                0.02490163230895996,
+                0.02484223937988281,
+                0.02470604705810547,
+                0.02471833610534668,
+                0.025219072341918947,
+                0.024705024719238283,
+                0.024656896591186524,
+                0.024748031616210937,
+                0.024853504180908204,
+                0.024802303314208983,
+                0.024672256469726563,
+                0.02469478416442871,
+                0.024624128341674805,
+                0.024782848358154298,
+                0.02466815948486328,
+                0.024713216781616212,
+                0.024809471130371095,
+                0.02485759925842285,
+                0.024619007110595705,
+                0.024747007369995116,
+                0.02472550392150879,
+                0.024773632049560547,
+                0.02464358329772949,
+                0.024666112899780275,
+                0.02476851272583008,
+                0.024778751373291014,
+                0.024756223678588866,
+                0.024809471130371095,
+                0.0246691837310791,
+                0.025225215911865235,
+                0.02474188804626465,
+                0.024887296676635744,
+                0.024833023071289064,
+                0.02496614456176758,
+                0.02472755241394043,
+                0.024805376052856445,
+                0.024830976486206056,
+                0.024859647750854492,
+                0.024780799865722656,
+                0.024855552673339845,
+                0.024794111251831053,
+                0.024844287872314453,
+                0.02474188804626465,
+                0.024704000473022462,
+                0.024803327560424804,
+                0.024780799865722656,
+                0.02471833610534668,
+                0.02474188804626465,
+                0.024653823852539062,
+                0.024818687438964843,
+                0.02467635154724121,
+                0.02469171142578125,
+                0.024634368896484377,
+                0.024688640594482423,
+                0.02467737579345703,
+                0.024695808410644532,
+                0.02474188804626465,
+                0.024788991928100586,
+                0.025223167419433593
             ]
         },
         "throughput": {
             "unit": "tokens/s",
+            "value": 40.38467204960459
         },
         "energy": null,
         "efficiency": null