diff --git "a/perf-df-gptq-1xT4.csv" "b/perf-df-gptq-1xT4.csv" --- "a/perf-df-gptq-1xT4.csv" +++ "b/perf-df-gptq-1xT4.csv" @@ -2387,7 +2387,7 @@ ChildProcessError: Traceback (most recent call last): RuntimeError: FlashAttention only supports Ampere GPUs or newer. ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v1-eager,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,qwen,Qwen/Qwen-7B,Qwen/Qwen-7B,cuda,0,42,,,True,True,,float16,True,False,,eager,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-eager,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,qwen,Qwen/Qwen-7B,Qwen/Qwen-7B,cuda,0,42,,,True,True,,float16,True,False,,eager,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -2839,7 +2839,7 @@ Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 4bit-gptq-exllama-v1-eager,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-1.3b,EleutherAI/pythia-1.3b,cuda,0,42,,,True,True,,float16,True,False,,eager,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.223-212.873.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.4.0,,4.44.2,,0.34.0,,,,1.21.4,,,,0.12.0,,,MB,1505.906688,1596.915712,0.0,1201.668096,1189.151232,s,1,8.179638671875,8.179638671875,0.0,8.179638671875,8.179638671875,8.179638671875,8.179638671875,[8.179638671875],,kWh,3.4542718758334936e-05,3.8032259343420668e-06,1.0888064266001071e-05,4.9234008958678075e-05,,MB,1524.883456,1791.950848,0.0,1382.023168,1351.367168,s,10,0.4748804168701172,0.04748804168701172,8.626458128418844e-05,0.04748456001281738,0.04758217430114746,0.04760303974151611,0.04761973209381103,"[0.04757753753662109, 0.04741312026977539, 0.047552288055419924, 0.04748134231567383, 0.04741155242919922, 0.04733055877685547, 0.04744131088256836, 0.047623905181884765, 0.04748777770996094, 0.0475610237121582]",tokens/s,5390.830847211323,kWh,1.4249498766260126e-06,1.5714683171962336e-07,9.500861259219454e-07,2.532182834267581e-06,tokens/kWh,101098544.91373903,MB,1528.918016,1833.893888,0.0,1423.966208,1407.328256,s,10,13.786848022460937,1.3786848022460938,0.0034189820094292227,1.3792277221679687,1.3835193969726562,1.3840247131347656,1.384428966064453,"[1.3790538330078126, 1.3803687744140625, 1.375017822265625, 1.3734044189453125, 1.379401611328125, 1.3756253662109375, 1.37638134765625, 1.3834071044921874, 1.37965771484375, 1.384530029296875]",tokens/s,45.695723850268834,kWh,4.049869935587363e-05,4.466583156182899e-06,1.8252484791678163e-05,6.32177673037347e-05,tokens/kWh,996555.2832214333,,s,630,13.784107973098747,0.021879536465236122,0.0002631747546613138,0.021808064460754395,0.022150712776184082,0.02236365451812744,0.022853109550476074,"[0.021778495788574218, 0.02174732780456543, 0.021688703536987305, 0.02183366394042969, 0.022048831939697266, 0.02204876708984375, 0.02185385513305664, 0.021788864135742186, 0.02177427291870117, 0.021960927963256837, 0.02182963180541992, 0.02185625648498535, 0.02177961540222168, 0.022281280517578127, 0.02176153564453125, 0.021694751739501954, 0.021700607299804688, 0.02172313690185547, 0.02201190376281738, 0.02191564750671387, 0.02171494483947754, 0.021778432846069336, 0.021729280471801758, 0.02169856071472168, 0.02166169548034668, 0.02194867134094238, 0.02190105628967285, 0.021749696731567382, 0.021851327896118163, 0.021825439453125, 0.02172003173828125, 0.021716991424560548, 0.021737472534179687, 0.02176924705505371, 0.021820383071899416, 0.021746784210205077, 0.021676895141601562, 0.021897279739379882, 0.021954559326171876, 0.022089824676513672, 0.022149023056030274, 0.022235136032104492, 0.02239190483093262, 0.022292959213256837, 0.022095680236816406, 0.022239328384399414, 0.0223045768737793, 0.02197987174987793, 0.022015071868896483, 0.022520767211914063, 0.02183718490600586, 0.0217872314453125, 0.02183884811401367, 0.02181427192687988, 0.021807104110717773, 0.02188902473449707, 0.021886144638061523, 0.021817983627319334, 0.021745311737060548, 0.021783136367797853, 0.021722272872924806, 0.02184419250488281, 0.021764671325683594, 0.021803680419921874, 0.021819391250610352, 0.021774240493774414, 0.021673919677734375, 0.021659807205200197, 0.021809152603149414, 0.021731327056884766, 0.021931392669677734, 0.021693056106567382, 0.021699583053588867, 0.021825824737548828, 0.021707487106323243, 0.021755456924438477, 0.022004159927368164, 0.022732223510742187, 0.02278438377380371, 0.021881023406982423, 0.022161407470703123, 0.021897216796875, 0.021882463455200195, 0.02214134407043457, 0.022901920318603514, 0.022288799285888672, 0.02204275131225586, 0.0219736328125, 0.02192918395996094, 0.021956607818603514, 0.02182192039489746, 0.022082624435424806, 0.02186502456665039, 0.021887487411499023, 0.02190652847290039, 0.021796735763549804, 0.02182636833190918, 0.021842016220092773, 0.021910720825195313, 0.02179078483581543, 0.022016767501831055, 0.021749536514282228, 0.021898527145385743, 0.02183468818664551, 0.021695552825927736, 0.021814207077026367, 0.02178656005859375, 0.021741024017333986, 0.021899871826171875, 0.021931264877319338, 0.021879552841186523, 0.021783552169799804, 0.021744640350341796, 0.021725183486938478, 0.02190745544433594, 0.022089471817016603, 0.02180940818786621, 0.021788671493530275, 0.02197283172607422, 0.021821599960327148, 0.021941791534423827, 0.021915935516357423, 0.021915199279785157, 0.021753856658935547, 0.021834239959716797, 0.021700735092163085, 0.02161664009094238, 0.02190745544433594, 0.02192793655395508, 0.021721088409423828, 0.021618015289306642, 0.02179715156555176, 0.02173334312438965, 0.022065568923950195, 0.021669792175292968, 0.021928031921386718, 0.021805055618286134, 0.02188083267211914, 0.022034431457519533, 0.021769535064697264, 0.021689023971557617, 0.02172867202758789, 0.02186240005493164, 0.02177276802062988, 0.02166592025756836, 0.021780384063720702, 0.02191574478149414, 0.021737472534179687, 0.021704704284667968, 0.021772064208984376, 0.02189334487915039, 0.022226943969726562, 0.02191321563720703, 0.021843807220458984, 0.021807199478149415, 0.02202684783935547, 0.021812448501586913, 0.021887615203857423, 0.021664928436279297, 0.02175881576538086, 0.02168320083618164, 0.02171494483947754, 0.0217807674407959, 0.021747936248779298, 0.021801279067993163, 0.021738719940185548, 0.021783424377441407, 0.021702783584594727, 0.021651424407958985, 0.021983232498168945, 0.021819391250610352, 0.02179452705383301, 0.021839391708374022, 0.021889184951782225, 0.022256223678588868, 0.021809152603149414, 0.02174729537963867, 0.021780672073364257, 0.021819616317749025, 0.021831872940063477, 0.021679935455322267, 0.021708799362182618, 0.0216760311126709, 0.022042240142822266, 0.02206528091430664, 0.021762304306030274, 0.021847711563110352, 0.021942655563354493, 0.021951839447021483, 0.022052671432495115, 0.02184796714782715, 0.02192207908630371, 0.0217161922454834, 0.021605152130126953, 0.021589632034301757, 0.0220546875, 0.021807712554931642, 0.021784223556518555, 0.021709152221679687, 0.02183737564086914, 0.022038976669311525, 0.021979135513305666, 0.021700191497802734, 0.021729503631591797, 0.021921503067016603, 0.021718591690063477, 0.02177935981750488, 0.021845407485961914, 0.02186204719543457, 0.021814207077026367, 0.021882015228271483, 0.021729536056518554, 0.02180134391784668, 0.021696735382080078, 0.021753856658935547, 0.021651456832885742, 0.021884735107421876, 0.021780351638793945, 0.021678176879882813, 0.021704927444458007, 0.021675296783447266, 0.021695104598999024, 0.021702560424804687, 0.021726879119873047, 0.02170854377746582, 0.021965599060058592, 0.02174787139892578, 0.021773439407348633, 0.022024927139282228, 0.021826976776123046, 0.02182819175720215, 0.02189427185058594, 0.0218919677734375, 0.021823488235473632, 0.021762048721313477, 0.021796863555908205, 0.021739519119262696, 0.021708383560180664, 0.021684576034545898, 0.02169196891784668, 0.02158355140686035, 0.02200454330444336, 0.021777952194213867, 0.02182806396484375, 0.021772287368774415, 0.02176425552368164, 0.021878623962402345, 0.021934080123901366, 0.021733375549316408, 0.02169036865234375, 0.02169254493713379, 0.021765888214111326, 0.021672607421875, 0.021807392120361327, 0.02176582336425781, 0.0217127685546875, 0.02174611282348633, 0.021827392578125, 0.022003807067871094, 0.02201113510131836, 0.021797727584838868, 0.022521535873413087, 0.021841279983520506, 0.021838560104370117, 0.021717023849487305, 0.02177039909362793, 0.021802400588989256, 0.02174425506591797, 0.021786624908447266, 0.021927679061889648, 0.021824928283691408, 0.021744480133056642, 0.022013887405395508, 0.02195167922973633, 0.021879167556762696, 0.021920255661010742, 0.02189030456542969, 0.02190822410583496, 0.021834943771362306, 0.021916479110717774, 0.021819391250610352, 0.02181875228881836, 0.021875328063964843, 0.02192153549194336, 0.02211164855957031, 0.021867359161376953, 0.02183123207092285, 0.021916095733642577, 0.021823488235473632, 0.021989280700683594, 0.02185635185241699, 0.021876735687255858, 0.02185420799255371, 0.021743583679199218, 0.02174569511413574, 0.02197020721435547, 0.021906143188476564, 0.021861440658569337, 0.021826240539550783, 0.021868799209594728, 0.02189030456542969, 0.021777280807495115, 0.021839744567871095, 0.021823488235473632, 0.021716928482055663, 0.021748960494995116, 0.02403619194030762, 0.021998624801635742, 0.021828575134277342, 0.021712928771972655, 0.021823455810546875, 0.021803007125854493, 0.021743616104125976, 0.022015775680541992, 0.021736831665039064, 0.0217893123626709, 0.021675519943237305, 0.021699071884155274, 0.021777664184570313, 0.021785343170166015, 0.021893119812011717, 0.02239676856994629, 0.02177155113220215, 0.02190835189819336, 0.02166774368286133, 0.021689855575561523, 0.02163158416748047, 0.021722944259643554, 0.021587392807006837, 0.021757728576660158, 0.021648319244384765, 0.021680160522460936, 0.021710048675537108, 0.022042943954467775, 0.021712352752685547, 0.02182601547241211, 0.021811456680297853, 0.021712736129760744, 0.021797632217407225, 0.021775136947631835, 0.021697248458862305, 0.021653696060180663, 0.02170675277709961, 0.0216760311126709, 0.021687583923339845, 0.021740255355834962, 0.021714559555053713, 0.021675615310668944, 0.021623296737670897, 0.022134559631347656, 0.02285532760620117, 0.02276438331604004, 0.02244607925415039, 0.02189107131958008, 0.0219965763092041, 0.021826528549194337, 0.021700607299804688, 0.021756128311157228, 0.022300447463989258, 0.021731327056884766, 0.02171683120727539, 0.0221812801361084, 0.021960447311401367, 0.021693376541137694, 0.021827648162841797, 0.021798912048339843, 0.021843584060668945, 0.021735807418823243, 0.021719039916992186, 0.021743616104125976, 0.021755903244018555, 0.021770240783691407, 0.021763711929321288, 0.02178291130065918, 0.021743616104125976, 0.021809152603149414, 0.02171452713012695, 0.021782943725585938, 0.021818944931030274, 0.02172153663635254, 0.021780479431152345, 0.021728927612304688, 0.02169215965270996, 0.021665599822998045, 0.021697311401367186, 0.021811199188232423, 0.021782527923583983, 0.021794815063476563, 0.021702655792236326, 0.021722944259643554, 0.02173766326904297, 0.021687519073486327, 0.021733631134033204, 0.021796640396118165, 0.021865215301513672, 0.021803007125854493, 0.021908863067626953, 0.021854240417480467, 0.021721696853637694, 0.021746847152709962, 0.02170163154602051, 0.021778207778930664, 0.021614431381225586, 0.021782720565795898, 0.021730527877807618, 0.022096607208251955, 0.021760095596313478, 0.021831680297851562, 0.021778079986572267, 0.02183407974243164, 0.021798816680908203, 0.02190985679626465, 0.02186809539794922, 0.0217458553314209, 0.021718271255493166, 0.02168294334411621, 0.021683616638183592, 0.021661951065063478, 0.021676383972167968, 0.02192140769958496, 0.02196019172668457, 0.022227807998657225, 0.0222761287689209, 0.022470304489135742, 0.022237184524536133, 0.022408639907836914, 0.021993663787841795, 0.021846912384033204, 0.02191958427429199, 0.021774335861206053, 0.02182713508605957, 0.021907039642333984, 0.021875551223754883, 0.022380544662475587, 0.021667295455932618, 0.02169705581665039, 0.02167955207824707, 0.021755552291870116, 0.02185513687133789, 0.021820608139038085, 0.022218656539916993, 0.021695552825927736, 0.021689056396484375, 0.02163430404663086, 0.02178553581237793, 0.021682207107543944, 0.021946367263793946, 0.022006944656372072, 0.022039392471313476, 0.021823488235473632, 0.021774335861206053, 0.021710847854614256, 0.021798912048339843, 0.02187468719482422, 0.021796319961547853, 0.022030624389648437, 0.02185241508483887, 0.021772287368774415, 0.021927871704101563, 0.021796384811401368, 0.021819328308105467, 0.02177084732055664, 0.02173855972290039, 0.022092832565307616, 0.02183353614807129, 0.02183900833129883, 0.021838399887084962, 0.021995904922485352, 0.02176963233947754, 0.021727872848510743, 0.021798879623413085, 0.021766271591186524, 0.022150463104248046, 0.021879360198974608, 0.021960704803466798, 0.02185830307006836, 0.02190745544433594, 0.021852224349975587, 0.021851104736328127, 0.022323200225830078, 0.021980127334594726, 0.02202828788757324, 0.022224704742431642, 0.0221529598236084, 0.022525503158569337, 0.022256511688232422, 0.021970943450927736, 0.021934080123901366, 0.02179180717468262, 0.021732383728027344, 0.021900800704956053, 0.02389232063293457, 0.023971904754638673, 0.022113407135009765, 0.021984128952026366, 0.02185215950012207, 0.021929983139038087, 0.021814815521240233, 0.021746143341064453, 0.02168832015991211, 0.021798303604125976, 0.02186835289001465, 0.02178656005859375, 0.021768224716186522, 0.021749343872070313, 0.0217542724609375, 0.021720127105712892, 0.02173014450073242, 0.02173347282409668, 0.021786624908447266, 0.021764095306396485, 0.02182713508605957, 0.02177609634399414, 0.02175049591064453, 0.02165555191040039, 0.021652544021606445, 0.021626976013183592, 0.021627744674682616, 0.021984512329101563, 0.02337420845031738, 0.022847679138183592, 0.022040735244750975, 0.022428800582885742, 0.021906335830688475, 0.021751808166503905, 0.021934080123901366, 0.02266659164428711, 0.02228700828552246, 0.02206105613708496, 0.022120288848876953, 0.022055072784423826, 0.021886816024780275, 0.02174991989135742, 0.02169990348815918, 0.021733184814453126, 0.021746559143066405, 0.021685407638549804, 0.021727775573730467, 0.02178041648864746, 0.02172496032714844, 0.021723583221435548, 0.02173513603210449, 0.021815296173095702, 0.02172140884399414, 0.02177142333984375, 0.02200054359436035, 0.02197715187072754, 0.021808416366577148, 0.021881568908691407, 0.021975040435791016, 0.022929407119750975, 0.021826751708984377, 0.021756736755371094, 0.021714015960693358, 0.021803680419921874, 0.021803264617919923, 0.021942272186279296, 0.02187059211730957, 0.022054912567138672, 0.021719039916992186, 0.021618688583374023, 0.021731199264526366, 0.021733503341674804, 0.02169206428527832, 0.02189961624145508, 0.021807104110717773, 0.02176598358154297, 0.02173967933654785, 0.021781984329223632, 0.02187913513183594, 0.021919551849365233, 0.021684608459472655, 0.02165555191040039, 0.021800960540771484, 0.021841920852661133, 0.02169343948364258, 0.02167091178894043, 0.021673759460449218, 0.021694719314575197, 0.021692384719848634, 0.02165113639831543, 0.021967168807983398, 0.021766143798828123, 0.021750848770141603, 0.02185100746154785, 0.021690208435058592, 0.021763904571533203, 0.021770656585693358, 0.021813247680664064, 0.02172313690185547, 0.02171494483947754, 0.021624704360961915, 0.021819360733032228, 0.021769792556762695, 0.02178927993774414, 0.02177340888977051, 0.021697439193725587, 0.02176348876953125, 0.02175030326843262, 0.021692480087280273, 0.021794815063476563, 0.021581823348999024, 0.02180019187927246, 0.021822208404541014, 0.021845184326171874, 0.02176083183288574, 0.021874015808105468, 0.021840543746948243, 0.022032384872436524, 0.02237824058532715, 0.02223708724975586, 0.022657567977905274, 0.022343488693237306, 0.022171648025512695, 0.02235935974121094, 0.02226655960083008, 0.022486080169677736, 0.02240812873840332, 0.02245631980895996, 0.022599679946899414, 0.022517759323120116, 0.02231888008117676, 0.022552799224853516, 0.02229043197631836, 0.02222198486328125, 0.022266719818115236, 0.022179840087890625, 0.022580480575561522, 0.02223904037475586, 0.02236716842651367]",tokens/s,45.704807393377656,, 4bit-gptq-exllama-v1-eager,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-1.4b,EleutherAI/pythia-1.4b,cuda,0,42,,,True,True,,float16,True,False,,eager,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.223-212.873.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.4.0,,4.44.2,,0.34.0,,,,1.21.4,,,,0.12.0,,,MB,1506.066432,1596.915712,0.0,1201.668096,1189.151232,s,1,8.4394453125,8.4394453125,0.0,8.4394453125,8.4394453125,8.4394453125,8.4394453125,[8.4394453125],,kWh,3.494024245833695e-05,3.846210831008193e-06,1.0811675316000496e-05,4.959812860534564e-05,,MB,1522.200576,1791.950848,0.0,1382.023168,1351.367168,s,10,0.4749305305480957,0.04749305305480957,0.00011023556915181127,0.04746963310241699,0.0475606086730957,0.047669617080688474,0.047756823806762695,"[0.04777862548828125, 0.047507423400878906, 0.04743027114868164, 0.047454689025878904, 0.047536384582519534, 0.04740451049804688, 0.04734524917602539, 0.047470401763916016, 0.047534111022949216, 0.04746886444091797]",tokens/s,5390.262017995812,kWh,1.4211896730987327e-06,1.5673211640795952e-07,9.467806926990326e-07,2.524702482205725e-06,tokens/kWh,101398086.23166707,MB,1527.451648,1833.893888,0.0,1423.966208,1407.328256,s,10,13.804923950195311,1.3804923950195311,0.005752157842506621,1.379092041015625,1.3883178833007812,1.3886030334472657,1.3888311535644533,"[1.377676513671875, 1.3759522705078124, 1.383525146484375, 1.3769593505859374, 1.3882545166015625, 1.3716900634765625, 1.37472216796875, 1.380507568359375, 1.38888818359375, 1.3867481689453125]",tokens/s,45.635890662844744,kWh,4.023062634273427e-05,4.4370357060596444e-06,1.8130345680100675e-05,6.27980077288946e-05,tokens/kWh,1003216.5394796827,,s,630,13.802479480743418,0.021908697588481598,0.00033537922035371343,0.021842127799987794,0.02214827880859375,0.02232315511703491,0.023321606845855717,"[0.02208777618408203, 0.022468511581420898, 0.021942304611206054, 0.02188287925720215, 0.021726303100585938, 0.021657855987548828, 0.021713184356689452, 0.021792896270751955, 0.021780736923217775, 0.021810176849365235, 0.021832704544067383, 0.021899295806884767, 0.021881887435913086, 0.021863359451293946, 0.02205695915222168, 0.02188902473449707, 0.021795936584472656, 0.021681055068969727, 0.02186854362487793, 0.021819391250610352, 0.021790271759033204, 0.021774784088134765, 0.021819263458251952, 0.02200752067565918, 0.02195088005065918, 0.021934080123901366, 0.021901311874389647, 0.02181679916381836, 0.02196329689025879, 0.021825216293334962, 0.021825632095336913, 0.02169264030456543, 0.021663679122924804, 0.021791967391967773, 0.021822303771972657, 0.021876735687255858, 0.0217227840423584, 0.021752511978149414, 0.021717920303344726, 0.021838336944580077, 0.021879039764404296, 0.022120447158813478, 0.021747711181640626, 0.02202828788757324, 0.0220446720123291, 0.02226380729675293, 0.0221529598236084, 0.02221696090698242, 0.02189673614501953, 0.021852863311767577, 0.021908319473266602, 0.021764448165893555, 0.02162505531311035, 0.021723520278930663, 0.021855552673339843, 0.02184262466430664, 0.02181724739074707, 0.0217109432220459, 0.021793920516967772, 0.021940223693847655, 0.021809535980224608, 0.021858495712280275, 0.021655296325683592, 0.021861215591430665, 0.0221265926361084, 0.02202828788757324, 0.021768192291259765, 0.021701856613159178, 0.02181353569030762, 0.021916160583496092, 0.0217969913482666, 0.02179574394226074, 0.02168931198120117, 0.02181065559387207, 0.022006303787231445, 0.021870304107666015, 0.021764127731323243, 0.02162441635131836, 0.021696928024291993, 0.021627071380615235, 0.021692480087280273, 0.021713056564331056, 0.021618816375732423, 0.02194112014770508, 0.02173401641845703, 0.021988639831542967, 0.02163580894470215, 0.0216529598236084, 0.021639904022216796, 0.021700159072875976, 0.021860832214355468, 0.02169856071472168, 0.021800960540771484, 0.02172047996520996, 0.021696128845214845, 0.0216560001373291, 0.021658143997192382, 0.0216944637298584, 0.02165113639831543, 0.021819007873535155, 0.021785280227661134, 0.02177142333984375, 0.021707584381103515, 0.021708768844604494, 0.02165510368347168, 0.021713312149047852, 0.02234582328796387, 0.025286848068237305, 0.021980607986450195, 0.021975488662719728, 0.021821151733398436, 0.021725439071655275, 0.021667999267578127, 0.021706911087036134, 0.021709983825683593, 0.021715456008911133, 0.021805055618286134, 0.02193017578125, 0.021912736892700194, 0.02188355255126953, 0.021760000228881835, 0.02176367950439453, 0.02175632095336914, 0.021768192291259765, 0.021776384353637695, 0.021628480911254883, 0.02236582374572754, 0.0216625919342041, 0.02176406478881836, 0.02185795211791992, 0.021551488876342773, 0.021654720306396483, 0.02156979179382324, 0.021749664306640625, 0.022313087463378907, 0.022829599380493164, 0.023365631103515624, 0.022073280334472655, 0.022161376953125, 0.022216064453125, 0.021853120803833007, 0.021657215118408204, 0.024447296142578127, 0.021847904205322264, 0.021849952697753906, 0.021932319641113283, 0.02182310485839844, 0.02184832000732422, 0.021929983139038087, 0.021878528594970702, 0.022477216720581054, 0.02195644760131836, 0.021795103073120117, 0.021894208908081053, 0.021889503479003907, 0.02184815979003906, 0.021881088256835938, 0.02190048027038574, 0.021815839767456054, 0.021929119110107424, 0.021950912475585938, 0.021988895416259764, 0.02203526306152344, 0.02194246482849121, 0.02191155242919922, 0.021777664184570313, 0.021861120223999022, 0.02188662338256836, 0.02180291175842285, 0.021757535934448242, 0.021876991271972655, 0.021860895156860353, 0.02191321563720703, 0.02187718391418457, 0.02206934356689453, 0.02184592056274414, 0.021780479431152345, 0.021735424041748046, 0.021750816345214842, 0.021758943557739257, 0.021957887649536132, 0.022002431869506837, 0.021825279235839844, 0.021740095138549805, 0.021677280426025392, 0.021670047760009765, 0.02187107276916504, 0.021782463073730468, 0.021798816680908203, 0.02171219253540039, 0.021844671249389647, 0.021613983154296874, 0.021748319625854492, 0.022103967666625975, 0.021784671783447264, 0.021767295837402344, 0.02174611282348633, 0.02170444869995117, 0.02181190490722656, 0.02183302307128906, 0.021783199310302735, 0.021693599700927733, 0.022080352783203125, 0.02197711944580078, 0.021842016220092773, 0.021736576080322267, 0.021826335906982422, 0.021841920852661133, 0.021825023651123047, 0.021694847106933594, 0.021784095764160155, 0.02175651168823242, 0.02183590316772461, 0.02175164794921875, 0.021927808761596678, 0.021698720932006837, 0.02195609664916992, 0.02197875213623047, 0.022218847274780275, 0.021961503982543946, 0.021954559326171876, 0.02182963180541992, 0.021722431182861327, 0.02176233673095703, 0.022051231384277344, 0.021921567916870117, 0.02176652717590332, 0.021864063262939454, 0.021719263076782225, 0.021855615615844728, 0.02209270477294922, 0.021880544662475587, 0.02178771209716797, 0.021922752380371092, 0.021782527923583983, 0.021762271881103516, 0.021923103332519532, 0.021899776458740236, 0.021854368209838868, 0.021895008087158205, 0.02183600044250488, 0.021774208068847656, 0.021790624618530274, 0.02175702476501465, 0.021885759353637697, 0.0219420166015625, 0.021987903594970704, 0.021851551055908202, 0.022118783950805663, 0.02193561553955078, 0.021934431076049806, 0.021827743530273436, 0.021836736679077148, 0.021729280471801758, 0.02169036865234375, 0.021800575256347657, 0.021739904403686523, 0.021817344665527344, 0.021940288543701173, 0.02198428726196289, 0.021869760513305664, 0.021978847503662108, 0.021869823455810546, 0.02172185516357422, 0.02179043197631836, 0.021817344665527344, 0.023213823318481444, 0.024300064086914062, 0.0221014404296875, 0.02187731170654297, 0.021934080123901366, 0.021737472534179687, 0.02176646423339844, 0.021939903259277343, 0.022360063552856444, 0.021862079620361328, 0.021866111755371093, 0.02308371162414551, 0.0220153923034668, 0.021895456314086913, 0.021962432861328124, 0.021844608306884766, 0.02180735969543457, 0.021839584350585937, 0.0218439998626709, 0.021925344467163085, 0.021821632385253906, 0.021754079818725586, 0.021723264694213866, 0.021786144256591797, 0.021811391830444334, 0.02175116729736328, 0.022369184494018556, 0.021955968856811524, 0.021814176559448242, 0.02185513687133789, 0.021918752670288085, 0.021806880950927733, 0.021917695999145507, 0.022011423110961915, 0.02187516784667969, 0.02182054328918457, 0.021808000564575197, 0.021806848526000976, 0.021857568740844727, 0.024050016403198243, 0.02310736083984375, 0.022021184921264647, 0.02186422348022461, 0.021796863555908205, 0.0218787841796875, 0.0220446720123291, 0.022828863143920897, 0.022120672225952147, 0.022045888900756837, 0.021937856674194334, 0.02186777687072754, 0.021787391662597657, 0.02161471939086914, 0.021600128173828125, 0.02168556785583496, 0.021665824890136718, 0.02172947120666504, 0.021600223541259764, 0.021586431503295898, 0.02182143974304199, 0.021936479568481444, 0.021799680709838867, 0.02166655921936035, 0.02157583999633789, 0.021661535263061523, 0.021710687637329102, 0.021893152236938475, 0.022117759704589842, 0.02195542335510254, 0.022003423690795897, 0.02239523124694824, 0.022083295822143554, 0.022098207473754884, 0.021929983139038087, 0.021894720077514647, 0.02172972869873047, 0.021757951736450197, 0.02180463981628418, 0.021903455734252928, 0.021848384857177734, 0.021663999557495116, 0.021595903396606445, 0.021675359725952147, 0.02171299171447754, 0.021713056564331056, 0.021596832275390623, 0.021626623153686523, 0.021722591400146485, 0.021776575088500977, 0.02185273551940918, 0.02182032012939453, 0.021738592147827147, 0.021759775161743163, 0.021716543197631836, 0.02167238426208496, 0.021725183486938478, 0.02162892723083496, 0.021689632415771484, 0.021662431716918944, 0.021785728454589842, 0.021618688583374023, 0.0216909122467041, 0.021723487854003905, 0.02169241523742676, 0.021769376754760743, 0.021821952819824218, 0.021748064041137695, 0.021811199188232423, 0.021670207977294922, 0.021669023513793944, 0.021658143997192382, 0.021792512893676758, 0.021791200637817382, 0.021731199264526366, 0.021686656951904297, 0.021786815643310548, 0.021659648895263672, 0.021618688583374023, 0.02180838394165039, 0.021828351974487306, 0.021901311874389647, 0.021702655792236326, 0.021694303512573242, 0.021785888671875, 0.021635711669921873, 0.021704959869384765, 0.021684288024902344, 0.02183558464050293, 0.02200726318359375, 0.022334112167358398, 0.021788768768310547, 0.02180415916442871, 0.021670623779296874, 0.021626752853393556, 0.02190905570983887, 0.021723743438720702, 0.02159619140625, 0.021794815063476563, 0.02186854362487793, 0.02175811195373535, 0.02173526382446289, 0.021790719985961913, 0.021937824249267577, 0.021840223312377928, 0.021713119506835937, 0.021724895477294923, 0.021868320465087892, 0.0218176326751709, 0.022011327743530273, 0.02206572723388672, 0.021842144012451173, 0.02194000053405762, 0.021929536819458008, 0.021920255661010742, 0.021747392654418947, 0.021852096557617186, 0.021974720001220704, 0.021913248062133787, 0.021808095932006836, 0.021724288940429687, 0.021955455780029297, 0.02176540756225586, 0.02180374336242676, 0.021784576416015625, 0.021720447540283204, 0.021792512893676758, 0.021817792892456056, 0.021710784912109374, 0.021664255142211913, 0.02181046485900879, 0.021875423431396486, 0.02223308753967285, 0.02190332794189453, 0.02196272087097168, 0.021780832290649414, 0.021789920806884765, 0.021792640686035158, 0.021898143768310546, 0.021855615615844728, 0.021670528411865234, 0.02163302421569824, 0.021752864837646484, 0.021871007919311524, 0.02182406425476074, 0.02178825569152832, 0.021668256759643553, 0.021816864013671874, 0.02180963134765625, 0.021902912139892577, 0.02167612838745117, 0.02163337516784668, 0.021744895935058593, 0.02172319984436035, 0.021834272384643555, 0.021811359405517577, 0.02180214309692383, 0.021840351104736328, 0.021862783432006837, 0.02191564750671387, 0.021812480926513673, 0.021850879669189454, 0.021948320388793945, 0.02190140724182129, 0.021876735687255858, 0.02195462417602539, 0.021839103698730468, 0.022005887985229493, 0.021907487869262696, 0.021976640701293945, 0.021804000854492186, 0.02209174346923828, 0.021980863571166992, 0.021970848083496093, 0.021933919906616212, 0.021910112380981447, 0.022138784408569336, 0.022306911468505858, 0.021975040435791016, 0.02230588722229004, 0.022006656646728514, 0.022063039779663087, 0.022029600143432616, 0.021865280151367187, 0.022585119247436523, 0.022034656524658202, 0.021929983139038087, 0.02188902473449707, 0.02185420799255371, 0.021772287368774415, 0.021792192459106446, 0.021955135345458985, 0.021921279907226563, 0.021891584396362306, 0.02186835289001465, 0.021932287216186522, 0.02189267158508301, 0.022294271469116212, 0.022198911666870116, 0.022417407989501953, 0.02215920066833496, 0.02197443199157715, 0.021887744903564453, 0.022714368820190428, 0.02211167907714844, 0.02228384017944336, 0.022207040786743164, 0.02214784049987793, 0.021814783096313475, 0.021842111587524415, 0.021934080123901366, 0.021989280700683594, 0.022206560134887695, 0.022039968490600585, 0.02193276786804199, 0.021874048233032226, 0.022082048416137694, 0.022331392288208008, 0.022171520233154298, 0.022017311096191407, 0.022227807998657225, 0.022124544143676757, 0.022050304412841795, 0.021873151779174805, 0.021927679061889648, 0.021909824371337892, 0.02182975959777832, 0.02187884712219238, 0.021843584060668945, 0.02179033660888672, 0.02174208068847656, 0.02173097610473633, 0.021801055908203124, 0.02180678367614746, 0.021709375381469727, 0.021747711181640626, 0.021784576416015625, 0.02186419105529785, 0.021989152908325194, 0.021915615081787108, 0.021774112701416017, 0.02176063919067383, 0.02172480010986328, 0.022883071899414062, 0.023624895095825195, 0.022219295501708983, 0.021976736068725584, 0.021887584686279295, 0.022031328201293946, 0.021953311920166016, 0.02196873664855957, 0.02239299201965332, 0.02224508857727051, 0.022255680084228516, 0.022353216171264647, 0.022068031311035158, 0.022126848220825196, 0.0221628475189209, 0.021832128524780274, 0.021816383361816405, 0.022019264221191406, 0.02187775993347168, 0.021813087463378907, 0.02181545639038086, 0.02178166389465332, 0.02215990447998047, 0.022110048294067382, 0.021965311050415038, 0.021794336318969727, 0.021868192672729492, 0.022483327865600585, 0.022421056747436524, 0.022393247604370118, 0.021807647705078124, 0.021798303604125976, 0.02170230484008789, 0.02195916748046875, 0.022182239532470702, 0.022289600372314453, 0.02194710350036621, 0.025202144622802736, 0.02198796844482422, 0.022403072357177735, 0.022409023284912108, 0.022170047760009765, 0.021812992095947267, 0.021729280471801758, 0.022263872146606446, 0.02225161552429199, 0.022152223587036134, 0.021957439422607423, 0.022183168411254884, 0.02192665672302246, 0.0218024959564209, 0.021822975158691405, 0.021716064453125, 0.02188035202026367, 0.021989503860473634, 0.02184976005554199, 0.022272319793701173, 0.021989856719970703, 0.022089536666870118, 0.021940223693847655, 0.021835775375366212, 0.02166988754272461, 0.021722400665283203, 0.021801824569702147, 0.021854080200195313, 0.02190336036682129, 0.02184601593017578, 0.021814367294311524, 0.021939104080200195, 0.02185420799255371, 0.021730655670166014, 0.021809375762939454, 0.02193174362182617, 0.021799583435058594, 0.021864511489868163, 0.021795040130615236, 0.021738431930541993, 0.02190598487854004, 0.022024192810058595, 0.02194486427307129, 0.02189695930480957, 0.021736928939819336]",tokens/s,45.64397294550935,, -4bit-gptq-exllama-v1-eager,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,,float16,True,False,,eager,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-eager,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,,float16,True,False,,eager,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -2884,7 +2884,7 @@ ChildProcessError: Traceback (most recent call last): self._buffers[key] = fn(buf) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1160, in convert return t.to( -torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 70.00 MiB. GPU 0 has a total capacity of 14.74 GiB of which 16.12 MiB is free. Process 19815 has 14.72 GiB memory in use. Of the allocated memory 14.44 GiB is allocated by PyTorch, and 187.67 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 70.00 MiB. GPU 0 has a total capacity of 14.74 GiB of which 16.12 MiB is free. Process 19654 has 14.72 GiB memory in use. Of the allocated memory 14.44 GiB is allocated by PyTorch, and 187.67 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 4bit-gptq-exllama-v1-eager,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-12b,EleutherAI/pythia-12b,cuda,0,42,,,True,True,,float16,True,False,,eager,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.223-212.873.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.4.0,,4.44.2,,0.34.0,,,,1.21.4,,,,0.12.0,,,MB,6768.16896,7762.542592,0.0,7367.294976,7351.94368,s,1,12.6473876953125,12.6473876953125,0.0,12.6473876953125,12.6473876953125,12.6473876953125,12.6473876953125,[12.6473876953125],,kWh,0.0001608182069083417,1.773204727626298e-05,4.962337303200323e-05,0.0002281736272166079,,MB,1658.08128,8372.813824,0.0,7962.886144,7872.44544,s,10,3.208184783935547,0.3208184783935547,0.0015482526784383604,0.3211701965332031,0.32249743347167964,0.322578581237793,0.3226434994506836,"[0.31727294921875, 0.3193846435546875, 0.3202622375488281, 0.3202374572753906, 0.32068597412109373, 0.3224794006347656, 0.3218287658691406, 0.3216544189453125, 0.32171920776367186, 0.32265972900390627]",tokens/s,797.9590243114346,kWh,9.335483573828066e-06,1.0295335762255684e-06,6.214831360749928e-06,1.6579848510803562e-05,tokens/kWh,15440430.582534475,MB,1666.674688,8624.472064,0.0,8214.544384,8118.577152,s,10,27.089406494140626,2.7089406494140627,0.004729172827181731,2.7089176025390627,2.7149712158203125,2.7152866333007815,2.7155389672851564,"[2.704410888671875, 2.714901123046875, 2.711612060546875, 2.701280029296875, 2.707578369140625, 2.70420947265625, 2.71560205078125, 2.7102568359375, 2.713653076171875, 2.705902587890625]",tokens/s,23.25632346859528,kWh,7.935913866033882e-05,8.753395499857032e-06,5.275871581805027e-05,0.00014087124997824615,tokens/kWh,447216.8736326872,,s,630,27.0860598335266,0.04299374576750256,0.0003475571523909005,0.042998958587646484,0.04339941749572754,0.04347417964935303,0.043852793464660644,"[0.043175136566162106, 0.04260927963256836, 0.042326175689697265, 0.042278911590576174, 0.04229935836791992, 0.04226233673095703, 0.04241388702392578, 0.04229721450805664, 0.04244534301757812, 0.04283391952514649, 0.04293603134155274, 0.04262704086303711, 0.04262508773803711, 0.042666240692138674, 0.04270489501953125, 0.042513664245605466, 0.04242918395996094, 0.042425472259521486, 0.04255424118041992, 0.04276627349853516, 0.043072895050048826, 0.04264400100708008, 0.04272947311401367, 0.04285164642333984, 0.042676544189453124, 0.042777118682861326, 0.04273299026489258, 0.042717758178710936, 0.0426673583984375, 0.04304553604125977, 0.04272127914428711, 0.04272281646728516, 0.04283030319213867, 0.04328656005859375, 0.04295091247558594, 0.04286233520507812, 0.042995712280273435, 0.04302627182006836, 0.04289539337158203, 0.04282102584838867, 0.04287561416625976, 0.042931617736816405, 0.043772350311279296, 0.04319452667236328, 0.04307891082763672, 0.043223583221435546, 0.04314470291137695, 0.043184864044189454, 0.04308582305908203, 0.04333884811401367, 0.04351084899902344, 0.04338876724243164, 0.04333878326416016, 0.04329776000976562, 0.043237377166748046, 0.043345375061035155, 0.04350620651245117, 0.04333977508544922, 0.04346060943603516, 0.043302913665771485, 0.0434952621459961, 0.043472705841064455, 0.043336032867431644, 0.043364864349365234, 0.042727294921875, 0.042464958190917966, 0.042301822662353515, 0.04230569458007812, 0.042411678314208986, 0.04323148727416992, 0.042772289276123046, 0.042735809326171874, 0.04303811264038086, 0.04293443298339844, 0.04286240005493164, 0.04269680023193359, 0.04265014266967773, 0.0425533447265625, 0.042906688690185546, 0.04272019195556641, 0.042668033599853515, 0.04272947311401367, 0.04288716888427734, 0.04304608154296875, 0.04324560165405274, 0.043031326293945314, 0.0432042236328125, 0.04300611114501953, 0.04313520050048828, 0.04304217529296875, 0.04284409713745117, 0.044810943603515625, 0.042923648834228514, 0.042874591827392575, 0.04308972930908203, 0.04332015991210938, 0.04316364669799805, 0.04313907241821289, 0.043199806213378905, 0.042832576751708984, 0.04298342514038086, 0.04301801681518555, 0.043191520690917966, 0.04295100784301758, 0.042984096527099606, 0.0432803840637207, 0.04321459197998047, 0.043235584259033205, 0.04326326370239258, 0.043324127197265624, 0.043549728393554685, 0.04349216079711914, 0.04337606430053711, 0.043301601409912106, 0.04330873489379883, 0.04346502304077148, 0.04342572784423828, 0.04344015884399414, 0.04330918502807617, 0.04337615966796875, 0.04341142272949219, 0.04343392181396484, 0.04336064147949219, 0.04335113525390625, 0.04332835388183594, 0.043323551177978516, 0.04378412628173828, 0.04307769775390625, 0.04273385620117188, 0.04267740631103516, 0.04379296112060547, 0.04258611297607422, 0.042510337829589843, 0.04253696060180664, 0.04282777786254883, 0.042802783966064455, 0.042844703674316406, 0.04269452667236328, 0.04265286254882812, 0.04262790298461914, 0.042447872161865234, 0.04236185455322266, 0.042792095184326175, 0.04294332885742187, 0.042979358673095706, 0.04304598236083984, 0.04305516815185547, 0.04295148849487305, 0.0432042236328125, 0.043106590270996094, 0.043219039916992184, 0.0431267204284668, 0.04299987030029297, 0.04300182342529297, 0.04302214431762695, 0.04272297668457031, 0.04273849487304687, 0.04298521423339844, 0.04286054229736328, 0.042727294921875, 0.04290070343017578, 0.04313910293579101, 0.04299660873413086, 0.042823680877685545, 0.042790912628173826, 0.04308377456665039, 0.04322742462158203, 0.04312998580932617, 0.0431129264831543, 0.043554943084716795, 0.04348435211181641, 0.043399295806884765, 0.043442878723144535, 0.04337254333496094, 0.04320665740966797, 0.043200511932373044, 0.04338278579711914, 0.04298076629638672, 0.04298179244995117, 0.04310444641113281, 0.043286720275878904, 0.043224895477294925, 0.04326358413696289, 0.0430838737487793, 0.04315343856811524, 0.04418588638305664, 0.04316774368286133, 0.043069438934326174, 0.04311859130859375, 0.04332729721069336, 0.0427760009765625, 0.042334720611572264, 0.04224227142333985, 0.04214505767822266, 0.04234400177001953, 0.042380287170410154, 0.04238131332397461, 0.042256385803222656, 0.04229529571533203, 0.04232585525512695, 0.042410144805908205, 0.04254265594482422, 0.042756542205810544, 0.04262911987304688, 0.04316998291015625, 0.04266374588012695, 0.042491233825683594, 0.04270662307739258, 0.04284515380859375, 0.04293001556396484, 0.04296720123291015, 0.04312678527832031, 0.043099838256835936, 0.04295507049560547, 0.043017822265625, 0.04270940780639648, 0.04268646240234375, 0.0426618881225586, 0.04274380874633789, 0.04292348861694336, 0.045267486572265626, 0.04263427352905273, 0.04264595031738281, 0.04264400100708008, 0.0425799674987793, 0.04271004867553711, 0.04305193710327149, 0.0428073616027832, 0.04279500961303711, 0.04286054229736328, 0.043222496032714844, 0.04295529556274414, 0.04301433563232422, 0.043036449432373045, 0.043216926574707035, 0.04325510406494141, 0.04335599899291992, 0.043103073120117186, 0.04330873489379883, 0.04307366561889649, 0.043159679412841795, 0.04308486557006836, 0.04294547271728515, 0.04293228912353515, 0.04294569778442383, 0.04314339065551758, 0.04308569717407226, 0.04297999954223633, 0.042998046875, 0.04310201644897461, 0.04306275177001953, 0.04313296127319336, 0.04342041778564453, 0.042923072814941406, 0.04268304061889648, 0.04245695877075195, 0.042523040771484374, 0.042502143859863284, 0.04253081512451172, 0.04278681564331055, 0.042659839630126956, 0.042567550659179686, 0.04259648132324219, 0.04259804916381836, 0.04273958587646484, 0.042856929779052734, 0.042649600982666014, 0.042641407012939454, 0.0430797119140625, 0.04279497528076172, 0.0428721923828125, 0.04324940872192383, 0.04305395126342773, 0.04312063980102539, 0.0432474250793457, 0.04321913528442383, 0.043122337341308596, 0.04296739196777344, 0.04294655990600586, 0.04298886489868164, 0.04289401626586914, 0.043014144897460936, 0.04270284652709961, 0.04286054229736328, 0.0427845458984375, 0.04271491241455078, 0.04285673522949219, 0.042626911163330075, 0.04286291122436523, 0.04267212677001953, 0.042883071899414066, 0.04324262237548828, 0.04294569778442383, 0.04293913650512695, 0.04305372619628906, 0.04340768051147461, 0.04332134246826172, 0.04333939361572266, 0.043321758270263674, 0.043337696075439455, 0.04309196853637695, 0.04382668685913086, 0.04306585693359375, 0.04290764617919922, 0.04312441635131836, 0.043477310180664065, 0.043245567321777346, 0.04310220718383789, 0.04324966430664062, 0.04322009658813476, 0.04309439849853516, 0.04290729522705078, 0.04290224075317383, 0.0431426887512207, 0.04330131149291992, 0.043869632720947266, 0.043270721435546874, 0.042651649475097655, 0.04259635162353516, 0.04260051345825195, 0.0426044807434082, 0.04250787353515625, 0.04245340728759766, 0.0425920639038086, 0.04257574462890625, 0.04250860977172852, 0.042684417724609375, 0.0425984001159668, 0.04245884704589844, 0.042473758697509766, 0.04240588760375977, 0.04229324722290039, 0.04250447845458984, 0.04273347091674805, 0.04280096054077148, 0.04284524917602539, 0.04294547271728515, 0.04292812728881836, 0.042928382873535155, 0.043020030975341794, 0.042979488372802736, 0.04280073547363281, 0.04271129608154297, 0.04288857650756836, 0.042961536407470705, 0.042638561248779294, 0.042549312591552736, 0.04258233642578125, 0.04249436950683594, 0.04265369415283203, 0.042676223754882815, 0.04270451354980469, 0.04268207931518555, 0.04268304061889648, 0.042706367492675784, 0.04297580718994141, 0.04300799942016602, 0.043044864654541014, 0.043243518829345705, 0.04332921600341797, 0.04344579315185547, 0.04337334442138672, 0.04329676818847656, 0.04333772659301758, 0.04314908981323242, 0.043308704376220704, 0.04311507034301758, 0.04339859390258789, 0.0432213134765625, 0.043151649475097656, 0.04324476623535156, 0.04338137435913086, 0.04332284927368164, 0.043380577087402346, 0.0433111686706543, 0.043299583435058596, 0.04343603134155274, 0.04347420883178711, 0.04339913558959961, 0.04303260803222656, 0.04261068725585938, 0.042665985107421874, 0.042510112762451174, 0.042610912322998046, 0.04250009536743164, 0.04269430541992188, 0.042690910339355466, 0.0426794548034668, 0.042686431884765626, 0.04272422409057617, 0.042659103393554686, 0.04273353576660156, 0.04263616180419922, 0.04261452865600586, 0.04288703918457031, 0.04281983947753906, 0.04295465469360352, 0.04302963256835937, 0.04312092971801758, 0.043067935943603516, 0.04314457702636719, 0.04344911956787109, 0.0433889274597168, 0.04321062469482422, 0.04319004821777344, 0.04307145690917969, 0.04307622528076172, 0.042833663940429687, 0.04305100631713867, 0.04298342514038086, 0.0429317741394043, 0.04286028671264648, 0.04282233428955078, 0.04310131072998047, 0.04322739028930664, 0.04371212768554687, 0.04313008117675781, 0.04309328079223633, 0.04327679824829102, 0.04336844635009766, 0.04324966430664062, 0.043227134704589845, 0.0434628791809082, 0.04340918350219727, 0.04336460876464844, 0.04359302520751953, 0.04339471817016602, 0.04343452835083008, 0.04332284927368164, 0.043485729217529294, 0.043612415313720704, 0.04323328018188476, 0.043361854553222653, 0.043153854370117185, 0.04337254333496094, 0.04334796905517578, 0.04329471969604492, 0.04329676818847656, 0.043466751098632815, 0.0434741439819336, 0.04342454528808594, 0.044015777587890624, 0.043071327209472654, 0.04274176025390625, 0.04261273574829102, 0.042592254638671875, 0.04266569519042969, 0.042514720916748044, 0.04254316711425781, 0.04259564971923828, 0.0426943359375, 0.04262499237060547, 0.042585247039794924, 0.04269622421264648, 0.04262659072875977, 0.042509056091308596, 0.043292671203613284, 0.04283596801757812, 0.04307558441162109, 0.042925216674804687, 0.04285116958618164, 0.04293632125854492, 0.04298150253295899, 0.04311846542358398, 0.04318767929077148, 0.04304540634155273, 0.04296499252319336, 0.0429439697265625, 0.04309657669067383, 0.0428787841796875, 0.04274812698364258, 0.04267827224731445, 0.04266921615600586, 0.04284288024902344, 0.042853759765625, 0.0429288330078125, 0.04280732727050781, 0.042731201171875, 0.042799488067626956, 0.04263724899291992, 0.042774528503417966, 0.043022335052490236, 0.04317388916015625, 0.04325580978393555, 0.04331433486938477, 0.04313993453979492, 0.04331699371337891, 0.04341785430908203, 0.043493377685546876, 0.04351795196533203, 0.043425792694091796, 0.043380062103271486, 0.043374591827392575, 0.044028575897216794, 0.043262046813964845, 0.04314720153808594, 0.04341574478149414, 0.04328015899658203, 0.04346265411376953, 0.04312998580932617, 0.0431396484375, 0.04314963150024414, 0.04316159820556641, 0.04323667144775391, 0.04386345672607422, 0.04311513519287109, 0.04253692626953125, 0.04272332763671875, 0.042768222808837894, 0.0427848014831543, 0.042829376220703125, 0.04252102279663086, 0.04254937744140625, 0.04265475082397461, 0.04252156829833984, 0.04261999893188476, 0.04249078369140625, 0.0427047348022461, 0.04272483062744141, 0.04268716812133789, 0.04311203384399414, 0.043108158111572266, 0.042885921478271485, 0.04316140747070313, 0.04310812759399414, 0.04309833526611328, 0.04340675354003906, 0.04357590484619141, 0.04315468978881836, 0.04315350341796875, 0.0431025276184082, 0.042889217376708984, 0.04286294555664062, 0.04276428985595703, 0.042773696899414064, 0.04284627151489258, 0.042910465240478514, 0.042882591247558596, 0.042931968688964844, 0.04295753479003906, 0.04287078475952148, 0.04290124893188477, 0.043014400482177736, 0.04313638305664062, 0.04307567977905274, 0.04341420745849609, 0.043764896392822265, 0.04334457778930664, 0.043402400970458985, 0.04345328140258789, 0.043361568450927736, 0.04321254348754883, 0.04325680160522461, 0.04315468978881836, 0.04310940933227539, 0.043173599243164065, 0.043200096130371096, 0.04331875228881836, 0.043138080596923825, 0.04319753646850586, 0.04346559906005859, 0.0434005126953125, 0.04327088165283203, 0.04349737548828125, 0.04348096084594726, 0.043432064056396484, 0.043490623474121096, 0.04354662322998047, 0.04286159896850586, 0.04266902542114258, 0.04239321517944336, 0.042377376556396486, 0.0426539192199707, 0.042573471069335934, 0.04252447891235352, 0.04252726364135742, 0.04244406509399414, 0.04257839965820313, 0.042493438720703124, 0.042558208465576175, 0.042543102264404296, 0.04253081512451172, 0.04278499221801758, 0.04280889511108398, 0.04281708908081055, 0.0427977294921875, 0.04288943862915039, 0.04315881729125977, 0.04314742279052734, 0.04312303924560547, 0.043089759826660155, 0.04301635360717773, 0.04305676651000977, 0.043032958984375, 0.04296089553833008, 0.04294620895385742, 0.04287932968139648, 0.04281327819824219, 0.042948257446289065, 0.04295731353759766, 0.04287078475952148, 0.042831871032714845, 0.0428361930847168, 0.04291766357421875, 0.04284633636474609, 0.0431962890625, 0.04320367813110351, 0.043170719146728515, 0.04303987121582031, 0.04301433563232422, 0.0432803840637207, 0.04325446319580078, 0.043128833770751954, 0.043096065521240234, 0.043093345642089845, 0.04324121475219726, 0.04332646560668945, 0.04329564666748047, 0.04312326431274414, 0.042932193756103514, 0.042936798095703124, 0.04297836685180664, 0.043021247863769534, 0.04305660629272461, 0.043409950256347654, 0.04320460891723633, 0.04338614273071289, 0.04315119934082031, 0.04305395126342773, 0.043186176300048826]",tokens/s,23.25919694012482,, @@ -2900,7 +2900,7 @@ torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 70.00 MiB. GPU 0 h raise RuntimeError(f""Isolated process exited with non-zero code {isolated_process.exitcode}"") RuntimeError: Isolated process exited with non-zero code -9 ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v1-eager,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,Deci/DeciCoder-1b,Deci/DeciCoder-1b,cuda,0,42,,,True,True,,float16,True,False,,eager,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-eager,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,Deci/DeciCoder-1b,Deci/DeciCoder-1b,cuda,0,42,,,True,True,,float16,True,False,,eager,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -3854,7 +3854,7 @@ ChildProcessError: Traceback (most recent call last): RuntimeError: FlashAttention only supports Ampere GPUs or newer. ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neo,EleutherAI/gpt-neo-2.7B,EleutherAI/gpt-neo-2.7B,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neo,EleutherAI/gpt-neo-2.7B,EleutherAI/gpt-neo-2.7B,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -4438,7 +4438,7 @@ ChildProcessError: Traceback (most recent call last): ValueError: RecurrentGemmaForCausalLM does not support Flash Attention 2.0 yet. Please request to add support where the model is hosted, on its model hub page: https://huggingface.co//tmp/tmpez1zazry/no_weights_model/discussions/new or in the Transformers GitHub repo: https://github.com/huggingface/transformers/issues/new ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gptj,EleutherAI/gpt-j-6b,EleutherAI/gpt-j-6b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gptj,EleutherAI/gpt-j-6b,EleutherAI/gpt-j-6b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -4765,7 +4765,7 @@ ChildProcessError: Traceback (most recent call last): torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 68.00 MiB. GPU 0 has a total capacity of 14.74 GiB of which 46.12 MiB is free. Process 57644 has 14.69 GiB memory in use. Of the allocated memory 14.29 GiB is allocated by PyTorch, and 313.73 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neo,EleutherAI/gpt-neo-125m,EleutherAI/gpt-neo-125m,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neo,EleutherAI/gpt-neo-125m,EleutherAI/gpt-neo-125m,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -4994,7 +4994,7 @@ ChildProcessError: Traceback (most recent call last): AttributeError: 'EmissionsTracker' object has no attribute '_scheduler' ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neo,EleutherAI/gpt-neo-1.3B,EleutherAI/gpt-neo-1.3B,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neo,EleutherAI/gpt-neo-1.3B,EleutherAI/gpt-neo-1.3B,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -5320,7 +5320,7 @@ ChildProcessError: Traceback (most recent call last): RuntimeError: FlashAttention only supports Ampere GPUs or newer. ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-1.3b,EleutherAI/pythia-1.3b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-1.3b,EleutherAI/pythia-1.3b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -5383,7 +5383,7 @@ ChildProcessError: Traceback (most recent call last): RuntimeError: FlashAttention only supports Ampere GPUs or newer. ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-1.4b,EleutherAI/pythia-1.4b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-1.4b,EleutherAI/pythia-1.4b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -5446,7 +5446,7 @@ ChildProcessError: Traceback (most recent call last): RuntimeError: FlashAttention only supports Ampere GPUs or newer. ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -5491,10 +5491,10 @@ ChildProcessError: Traceback (most recent call last): self._buffers[key] = fn(buf) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1160, in convert return t.to( -torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 70.00 MiB. GPU 0 has a total capacity of 14.74 GiB of which 16.12 MiB is free. Process 22554 has 14.72 GiB memory in use. Of the allocated memory 14.44 GiB is allocated by PyTorch, and 187.67 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 70.00 MiB. GPU 0 has a total capacity of 14.74 GiB of which 16.12 MiB is free. Process 22577 has 14.72 GiB memory in use. Of the allocated memory 14.44 GiB is allocated by PyTorch, and 187.67 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-12b,EleutherAI/pythia-12b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-12b,EleutherAI/pythia-12b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -5699,7 +5699,7 @@ RuntimeError: FlashAttention only supports Ampere GPUs or newer. raise RuntimeError(f""Isolated process exited with non-zero code {isolated_process.exitcode}"") RuntimeError: Isolated process exited with non-zero code -9 ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,Deci/DeciCoder-1b,Deci/DeciCoder-1b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,Deci/DeciCoder-1b,Deci/DeciCoder-1b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -6206,7 +6206,7 @@ ChildProcessError: Traceback (most recent call last): ValueError: RecurrentGemmaForCausalLM does not support Flash Attention 2.0 yet. Please request to add support where the model is hosted, on its model hub page: https://huggingface.co//tmp/tmpq_8o8mfj/no_weights_model/discussions/new or in the Transformers GitHub repo: https://github.com/huggingface/transformers/issues/new ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/gpt-neox-20b,EleutherAI/gpt-neox-20b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/gpt-neox-20b,EleutherAI/gpt-neox-20b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -6278,7 +6278,7 @@ RuntimeError: FlashAttention only supports Ampere GPUs or newer. raise RuntimeError(f""Isolated process exited with non-zero code {isolated_process.exitcode}"") RuntimeError: Isolated process exited with non-zero code -9 ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,01-ai/Yi-6B,01-ai/Yi-6B,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,01-ai/Yi-6B,01-ai/Yi-6B,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -6404,7 +6404,7 @@ ChildProcessError: Traceback (most recent call last): RuntimeError: FlashAttention only supports Ampere GPUs or newer. ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-160m,EleutherAI/pythia-160m,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-160m,EleutherAI/pythia-160m,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -6467,7 +6467,7 @@ ChildProcessError: Traceback (most recent call last): RuntimeError: FlashAttention only supports Ampere GPUs or newer. ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-6.7b,EleutherAI/pythia-6.7b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-6.7b,EleutherAI/pythia-6.7b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -6719,7 +6719,7 @@ ChildProcessError: Traceback (most recent call last): RuntimeError: FlashAttention only supports Ampere GPUs or newer. ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-2.7b,EleutherAI/pythia-2.7b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-2.7b,EleutherAI/pythia-2.7b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -6917,7 +6917,7 @@ ChildProcessError: Traceback (most recent call last): RuntimeError: FlashAttention only supports Ampere GPUs or newer. ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-70m,EleutherAI/pythia-70m,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-70m,EleutherAI/pythia-70m,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -7369,7 +7369,7 @@ ChildProcessError: Traceback (most recent call last): RuntimeError: FlashAttention only supports Ampere GPUs or newer. ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/polyglot-ko-12.8b,EleutherAI/polyglot-ko-12.8b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/polyglot-ko-12.8b,EleutherAI/polyglot-ko-12.8b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -7530,7 +7530,7 @@ ChildProcessError: Traceback (most recent call last): ValueError: CodeGenForCausalLM does not support Flash Attention 2.0 yet. Please request to add support where the model is hosted, on its model hub page: https://huggingface.co//tmp/tmpgpbskzdx/no_weights_model/discussions/new or in the Transformers GitHub repo: https://github.com/huggingface/transformers/issues/new ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-410m,EleutherAI/pythia-410m,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v2-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-410m,EleutherAI/pythia-410m,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -8144,7 +8144,7 @@ ChildProcessError: Traceback (most recent call last): RuntimeError: FlashAttention only supports Ampere GPUs or newer. ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neo,EleutherAI/gpt-neo-2.7B,EleutherAI/gpt-neo-2.7B,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neo,EleutherAI/gpt-neo-2.7B,EleutherAI/gpt-neo-2.7B,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -8728,7 +8728,7 @@ ChildProcessError: Traceback (most recent call last): ValueError: RecurrentGemmaForCausalLM does not support Flash Attention 2.0 yet. Please request to add support where the model is hosted, on its model hub page: https://huggingface.co//tmp/tmpp70zzfht/no_weights_model/discussions/new or in the Transformers GitHub repo: https://github.com/huggingface/transformers/issues/new ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gptj,EleutherAI/gpt-j-6b,EleutherAI/gpt-j-6b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gptj,EleutherAI/gpt-j-6b,EleutherAI/gpt-j-6b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -9055,7 +9055,7 @@ ChildProcessError: Traceback (most recent call last): torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 68.00 MiB. GPU 0 has a total capacity of 14.74 GiB of which 46.12 MiB is free. Process 57020 has 14.69 GiB memory in use. Of the allocated memory 14.29 GiB is allocated by PyTorch, and 313.73 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neo,EleutherAI/gpt-neo-125m,EleutherAI/gpt-neo-125m,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neo,EleutherAI/gpt-neo-125m,EleutherAI/gpt-neo-125m,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -9284,7 +9284,7 @@ ChildProcessError: Traceback (most recent call last): AttributeError: 'EmissionsTracker' object has no attribute '_scheduler' ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neo,EleutherAI/gpt-neo-1.3B,EleutherAI/gpt-neo-1.3B,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neo,EleutherAI/gpt-neo-1.3B,EleutherAI/gpt-neo-1.3B,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -9610,7 +9610,7 @@ ChildProcessError: Traceback (most recent call last): RuntimeError: FlashAttention only supports Ampere GPUs or newer. ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-1.3b,EleutherAI/pythia-1.3b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-1.3b,EleutherAI/pythia-1.3b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -9673,7 +9673,7 @@ ChildProcessError: Traceback (most recent call last): RuntimeError: FlashAttention only supports Ampere GPUs or newer. ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-1.4b,EleutherAI/pythia-1.4b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-1.4b,EleutherAI/pythia-1.4b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -9736,7 +9736,7 @@ ChildProcessError: Traceback (most recent call last): RuntimeError: FlashAttention only supports Ampere GPUs or newer. ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -9781,10 +9781,10 @@ ChildProcessError: Traceback (most recent call last): self._buffers[key] = fn(buf) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1160, in convert return t.to( -torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 70.00 MiB. GPU 0 has a total capacity of 14.74 GiB of which 16.12 MiB is free. Process 21977 has 14.72 GiB memory in use. Of the allocated memory 14.44 GiB is allocated by PyTorch, and 187.67 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 70.00 MiB. GPU 0 has a total capacity of 14.74 GiB of which 16.12 MiB is free. Process 22037 has 14.72 GiB memory in use. Of the allocated memory 14.44 GiB is allocated by PyTorch, and 187.67 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-12b,EleutherAI/pythia-12b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-12b,EleutherAI/pythia-12b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -9989,7 +9989,7 @@ RuntimeError: FlashAttention only supports Ampere GPUs or newer. raise RuntimeError(f""Isolated process exited with non-zero code {isolated_process.exitcode}"") RuntimeError: Isolated process exited with non-zero code -9 ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,Deci/DeciCoder-1b,Deci/DeciCoder-1b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,Deci/DeciCoder-1b,Deci/DeciCoder-1b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -10496,7 +10496,7 @@ ChildProcessError: Traceback (most recent call last): ValueError: RecurrentGemmaForCausalLM does not support Flash Attention 2.0 yet. Please request to add support where the model is hosted, on its model hub page: https://huggingface.co//tmp/tmpnx61_yf_/no_weights_model/discussions/new or in the Transformers GitHub repo: https://github.com/huggingface/transformers/issues/new ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/gpt-neox-20b,EleutherAI/gpt-neox-20b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/gpt-neox-20b,EleutherAI/gpt-neox-20b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -10568,7 +10568,7 @@ RuntimeError: FlashAttention only supports Ampere GPUs or newer. raise RuntimeError(f""Isolated process exited with non-zero code {isolated_process.exitcode}"") RuntimeError: Isolated process exited with non-zero code -9 ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,01-ai/Yi-6B,01-ai/Yi-6B,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,01-ai/Yi-6B,01-ai/Yi-6B,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -10694,7 +10694,7 @@ ChildProcessError: Traceback (most recent call last): RuntimeError: FlashAttention only supports Ampere GPUs or newer. ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-160m,EleutherAI/pythia-160m,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-160m,EleutherAI/pythia-160m,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -10757,7 +10757,7 @@ ChildProcessError: Traceback (most recent call last): RuntimeError: FlashAttention only supports Ampere GPUs or newer. ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-6.7b,EleutherAI/pythia-6.7b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-6.7b,EleutherAI/pythia-6.7b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -11009,7 +11009,7 @@ ChildProcessError: Traceback (most recent call last): RuntimeError: FlashAttention only supports Ampere GPUs or newer. ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-2.7b,EleutherAI/pythia-2.7b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-2.7b,EleutherAI/pythia-2.7b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -11207,7 +11207,7 @@ ChildProcessError: Traceback (most recent call last): RuntimeError: FlashAttention only supports Ampere GPUs or newer. ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-70m,EleutherAI/pythia-70m,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-70m,EleutherAI/pythia-70m,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -11659,7 +11659,7 @@ ChildProcessError: Traceback (most recent call last): RuntimeError: FlashAttention only supports Ampere GPUs or newer. ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/polyglot-ko-12.8b,EleutherAI/polyglot-ko-12.8b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/polyglot-ko-12.8b,EleutherAI/polyglot-ko-12.8b,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -11820,7 +11820,7 @@ ChildProcessError: Traceback (most recent call last): ValueError: CodeGenForCausalLM does not support Flash Attention 2.0 yet. Please request to add support where the model is hosted, on its model hub page: https://huggingface.co//tmp/tmprhv7knqk/no_weights_model/discussions/new or in the Transformers GitHub repo: https://github.com/huggingface/transformers/issues/new ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-410m,EleutherAI/pythia-410m,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-flash_attention_2,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-410m,EleutherAI/pythia-410m,cuda,0,42,,,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -12152,7 +12152,7 @@ ValueError: OPTForCausalLM does not support an attention implementation through ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 4bit-gptq-exllama-v1-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,stabilityai/stablelm-base-alpha-7b,stabilityai/stablelm-base-alpha-7b,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,,MB,4896.751616,5995.626496,0.0,5593.104384,5582.857216,s,1,11.489771484375,11.489771484375,0.0,11.489771484375,11.489771484375,11.489771484375,11.489771484375,[11.489771484375],,kWh,0.00011806145224998848,1.301551743460821e-05,3.635197352599984e-05,0.00016742894321059652,,MB,1569.632256,6324.87936,0.0,5907.677184,5844.559872,s,10,2.0207774047851563,0.20207774047851562,0.00038689895244857,0.20201309967041015,0.202544287109375,0.20273568267822267,0.2028887991333008,"[0.20195497131347656, 0.20207122802734376, 0.2018753662109375, 0.2016386260986328, 0.2021746826171875, 0.2017010498046875, 0.2016741485595703, 0.20292707824707032, 0.2022584991455078, 0.2025017547607422]",tokens/s,1266.8391847305777,kWh,5.911609283416207e-06,6.519431479882893e-07,3.9273975863601815e-06,1.0490950017764678e-05,tokens/kWh,24401984.526330467,MB,1573.670912,6324.87936,0.0,5907.677184,5844.562432,s,10,14.907234375,1.4907234375,0.0018872144851043732,1.4902535400390624,1.4928083984375,1.4931936889648436,1.4935019213867187,"[1.4899151611328125, 1.4897984619140625, 1.489876953125, 1.4917333984375, 1.4927227783203125, 1.4935789794921874, 1.489792236328125, 1.4867083740234375, 1.49251611328125, 1.4905919189453125]",tokens/s,42.261360098861395,kWh,4.35392782282485e-05,4.802132315365114e-06,2.8924045361438113e-05,7.726545590505172e-05,tokens/kWh,815370.8440861083,,s,630,14.904556339263905,0.023658025935339548,0.00018368785427193713,0.02365676784515381,0.023894118690490723,0.02396721887588501,0.024107075710296633,"[0.023576576232910155, 0.023207935333251953, 0.02334048080444336, 0.02360697555541992, 0.02399724769592285, 0.023935264587402343, 0.02375657653808594, 0.02368307113647461, 0.023910400390625, 0.023704959869384767, 0.02363216018676758, 0.023538015365600587, 0.023580671310424805, 0.023748607635498048, 0.02356547164916992, 0.0237076473236084, 0.02373472023010254, 0.023689504623413085, 0.02355628776550293, 0.02344691276550293, 0.023321151733398438, 0.02345779228210449, 0.0239554557800293, 0.023750656127929686, 0.02369945526123047, 0.023673887252807616, 0.023533920288085937, 0.02353011131286621, 0.02368307113647461, 0.023735647201538087, 0.023720703125, 0.02387958335876465, 0.023602176666259765, 0.023791648864746093, 0.023808992385864258, 0.02386534309387207, 0.023525568008422853, 0.023442752838134767, 0.023441919326782225, 0.023572288513183593, 0.02374675178527832, 0.023776384353637697, 0.023645055770874022, 0.023610815048217774, 0.023409215927124024, 0.02342092704772949, 0.02353683280944824, 0.023660991668701174, 0.02365603256225586, 0.023589664459228516, 0.023502656936645508, 0.02356038475036621, 0.02367897605895996, 0.023608608245849608, 0.02356502342224121, 0.023529472351074218, 0.02345062446594238, 0.02376192092895508, 0.02389401626586914, 0.02381814384460449, 0.02383395195007324, 0.02375913619995117, 0.023752544403076174, 0.023408479690551758, 0.023318080902099608, 0.023470304489135743, 0.02357491111755371, 0.023391904830932616, 0.02329635238647461, 0.023824384689331055, 0.02391859245300293, 0.02371174430847168, 0.023599103927612306, 0.023576543807983397, 0.02327334403991699, 0.023424575805664063, 0.023511648178100586, 0.023451648712158202, 0.02343731117248535, 0.0233984317779541, 0.023286975860595704, 0.023364383697509764, 0.023556095123291015, 0.023492319107055664, 0.023716127395629883, 0.023713983535766602, 0.023445344924926757, 0.02357449531555176, 0.023492000579833985, 0.02367958450317383, 0.023654399871826173, 0.023597055435180665, 0.023576576232910155, 0.023737375259399413, 0.02380080032348633, 0.02396073532104492, 0.02388380813598633, 0.023689311981201173, 0.023694047927856444, 0.023545856475830077, 0.023508991241455078, 0.02353104019165039, 0.023467615127563478, 0.023843103408813477, 0.02374838447570801, 0.02414620780944824, 0.024066688537597657, 0.023996320724487305, 0.023787200927734373, 0.023590463638305664, 0.02361952018737793, 0.023737152099609374, 0.023734399795532227, 0.02367840003967285, 0.023583168029785155, 0.023513088226318358, 0.02368307113647461, 0.023832576751708984, 0.024330015182495116, 0.023812320709228514, 0.023752704620361328, 0.023495872497558593, 0.02373302459716797, 0.023773216247558595, 0.023812095642089845, 0.023725791931152342, 0.023531295776367187, 0.02323036766052246, 0.02351103973388672, 0.023646175384521486, 0.023626943588256837, 0.023640607833862303, 0.023376415252685547, 0.02357372856140137, 0.023417407989501954, 0.02332262420654297, 0.02355990409851074, 0.023421119689941407, 0.023568479537963868, 0.023684127807617188, 0.023690080642700194, 0.02376905632019043, 0.02374774360656738, 0.023634944915771484, 0.02345302391052246, 0.023640735626220703, 0.023554048538208007, 0.023441408157348635, 0.023536960601806642, 0.02342540740966797, 0.023487808227539063, 0.023524351119995117, 0.023711551666259767, 0.023787551879882813, 0.02381020736694336, 0.023756223678588866, 0.023571008682250975, 0.02372812843322754, 0.023686912536621092, 0.02380620765686035, 0.023643680572509766, 0.02348480033874512, 0.023365728378295897, 0.023457151412963867, 0.023693695068359374, 0.02368307113647461, 0.023799135208129884, 0.0236430721282959, 0.023502815246582032, 0.02370969581604004, 0.023598335266113282, 0.023618303298950195, 0.023609344482421874, 0.02371993637084961, 0.023613439559936524, 0.023751840591430665, 0.024072608947753905, 0.023971712112426758, 0.023940736770629883, 0.02377129554748535, 0.0236080322265625, 0.023588960647583007, 0.024001728057861327, 0.02404377555847168, 0.02387171173095703, 0.023732383728027343, 0.02359926414489746, 0.023773248672485352, 0.02384480094909668, 0.023587039947509766, 0.023258047103881838, 0.023411680221557617, 0.023672384262084963, 0.024004480361938477, 0.023686975479125977, 0.02374131202697754, 0.023698400497436524, 0.023634016036987306, 0.023614208221435548, 0.023731391906738283, 0.02356928062438965, 0.023607295989990236, 0.023587072372436523, 0.023755680084228514, 0.023382879257202147, 0.023508224487304687, 0.023520000457763673, 0.023427072525024413, 0.023287647247314452, 0.023198944091796875, 0.023374784469604493, 0.02350668716430664, 0.0235513916015625, 0.023595392227172853, 0.023463647842407228, 0.02333158493041992, 0.023371776580810546, 0.02365388870239258, 0.02386956787109375, 0.02384499168395996, 0.0237521915435791, 0.02375142478942871, 0.023848575592041017, 0.02409721565246582, 0.02407580757141113, 0.023939039230346678, 0.023730112075805665, 0.02348297691345215, 0.023494560241699217, 0.02368921661376953, 0.023784608840942384, 0.023726335525512697, 0.023851871490478516, 0.023561983108520507, 0.02369740867614746, 0.023863040924072265, 0.023619104385375976, 0.023857919692993165, 0.023446752548217775, 0.023476991653442383, 0.0240960636138916, 0.02411494445800781, 0.02418783950805664, 0.023828191757202147, 0.0236014404296875, 0.02362544059753418, 0.023975711822509765, 0.02402355194091797, 0.023940895080566408, 0.02368534469604492, 0.02352499198913574, 0.02365683174133301, 0.02339619255065918, 0.023602943420410156, 0.023610559463500977, 0.023555904388427733, 0.023687231063842774, 0.02374239921569824, 0.023895488739013673, 0.023663167953491212, 0.02366806411743164, 0.023624351501464844, 0.02362508773803711, 0.023626367568969728, 0.023643999099731444, 0.023678783416748048, 0.023562591552734376, 0.02332057571411133, 0.023351264953613282, 0.023592031478881836, 0.02361235237121582, 0.023624799728393556, 0.023538591384887696, 0.02349171257019043, 0.023632768630981446, 0.023802911758422852, 0.023896831512451172, 0.023752927780151367, 0.02373017692565918, 0.023650304794311523, 0.02367283248901367, 0.02378550338745117, 0.023934112548828126, 0.02393974494934082, 0.023850719451904298, 0.023677568435668945, 0.02385055923461914, 0.023861503601074217, 0.02384432029724121, 0.023700000762939454, 0.02350694465637207, 0.02345523262023926, 0.023640575408935546, 0.023719711303710936, 0.023775455474853515, 0.023612960815429688, 0.02366476821899414, 0.023755104064941406, 0.023785472869873047, 0.023846912384033202, 0.023769088745117187, 0.023565759658813478, 0.02344361686706543, 0.023599519729614257, 0.02411110305786133, 0.024037439346313475, 0.023860607147216797, 0.023931455612182618, 0.02354380798339844, 0.023643583297729493, 0.023654975891113282, 0.023775232315063476, 0.023820287704467775, 0.023694751739501953, 0.023551679611206053, 0.023560224533081056, 0.023494495391845702, 0.023667903900146486, 0.023646848678588867, 0.02350726318359375, 0.023557119369506836, 0.023596031188964844, 0.02357593536376953, 0.02360697555541992, 0.023491519927978516, 0.02343065643310547, 0.023306751251220705, 0.023385503768920898, 0.02357088088989258, 0.023548095703125, 0.02354172706604004, 0.023603391647338868, 0.02325801658630371, 0.02342799949645996, 0.023533567428588868, 0.023601152420043944, 0.023643775939941405, 0.023667327880859373, 0.023527168273925782, 0.023629823684692384, 0.023820287704467775, 0.0237806396484375, 0.023797855377197266, 0.02386944007873535, 0.023437952041625975, 0.023666688919067383, 0.02395955276489258, 0.024007936477661133, 0.023803712844848633, 0.02403424072265625, 0.02385305595397949, 0.02389504051208496, 0.023778303146362305, 0.02369536018371582, 0.02373206329345703, 0.02364348793029785, 0.023671648025512696, 0.024153440475463868, 0.023961471557617187, 0.02376780891418457, 0.02375587272644043, 0.023571359634399415, 0.023812095642089845, 0.023948448181152344, 0.0239215030670166, 0.023811904907226563, 0.023726272583007812, 0.023680288314819335, 0.023768800735473633, 0.024005632400512695, 0.023958623886108397, 0.02378201675415039, 0.023687456130981447, 0.023769088745117187, 0.023900159835815428, 0.023820032119750978, 0.023895360946655272, 0.023763200759887696, 0.023801376342773437, 0.023770048141479493, 0.023744512557983398, 0.02366464042663574, 0.023577760696411133, 0.02350166320800781, 0.02345680046081543, 0.023436256408691406, 0.023412607192993165, 0.023404447555541993, 0.0235828800201416, 0.02341257667541504, 0.023273120880126952, 0.023320959091186522, 0.023463775634765625, 0.023492639541625976, 0.023518720626831056, 0.023477119445800783, 0.02335532760620117, 0.023318656921386717, 0.02356755256652832, 0.0238720645904541, 0.023729728698730468, 0.023740991592407227, 0.023615488052368162, 0.023259136199951173, 0.02352275276184082, 0.023708223342895508, 0.023738208770751952, 0.023674367904663086, 0.023454368591308592, 0.023649280548095702, 0.023814271926879883, 0.023947551727294923, 0.023924512863159178, 0.023825151443481445, 0.023684703826904296, 0.02358278465270996, 0.02374083137512207, 0.023756479263305662, 0.023734592437744142, 0.023627775192260742, 0.02346294403076172, 0.023917024612426757, 0.023839231491088866, 0.023803295135498045, 0.023722591400146483, 0.023504896163940428, 0.023529472351074218, 0.0235599365234375, 0.02373868751525879, 0.023700447082519532, 0.023896831512451172, 0.02376246452331543, 0.02366681671142578, 0.02367340850830078, 0.023795040130615234, 0.023749216079711914, 0.023729888916015626, 0.023497055053710938, 0.023762752532958984, 0.02391904067993164, 0.024090368270874022, 0.023929920196533203, 0.02326905632019043, 0.023480543136596678, 0.023793664932250977, 0.023756160736083984, 0.023693952560424805, 0.023605247497558594, 0.023586816787719726, 0.02357369613647461, 0.02365523147583008, 0.023568384170532225, 0.023670656204223633, 0.023533695220947264, 0.023396352767944335, 0.023339008331298827, 0.023391679763793947, 0.02332320022583008, 0.023259136199951173, 0.02326937675476074, 0.02340838432312012, 0.02343756866455078, 0.023438880920410157, 0.02344598388671875, 0.02343731117248535, 0.02367692756652832, 0.023529472351074218, 0.023414783477783203, 0.023408639907836915, 0.023455743789672853, 0.02367692756652832, 0.023609024047851562, 0.023438880920410157, 0.02345619201660156, 0.02363382339477539, 0.02372428894042969, 0.023636224746704102, 0.023495935440063478, 0.023480512619018554, 0.023494783401489257, 0.02361577606201172, 0.023591007232666016, 0.023656160354614257, 0.023634464263916015, 0.023627519607543945, 0.023813728332519532, 0.023868928909301756, 0.023855615615844726, 0.023855520248413087, 0.02367487907409668, 0.023541759490966797, 0.0237260799407959, 0.02368716812133789, 0.023711584091186524, 0.02372822380065918, 0.023773248672485352, 0.023472127914428712, 0.023563743591308594, 0.02376144027709961, 0.02370560073852539, 0.023752704620361328, 0.02376857566833496, 0.02366105651855469, 0.024012672424316407, 0.02360940742492676, 0.023428800582885743, 0.02343142318725586, 0.02332806396484375, 0.023451679229736327, 0.02354256057739258, 0.023711200714111327, 0.02380611228942871, 0.023752960205078125, 0.02372198486328125, 0.023746559143066406, 0.02352479934692383, 0.023683647155761718, 0.02372812843322754, 0.023586528778076172, 0.023510719299316408, 0.02380771255493164, 0.023947519302368166, 0.02388582420349121, 0.023682815551757812, 0.023610240936279298, 0.02349260711669922, 0.023619583129882812, 0.02380739212036133, 0.023754911422729494, 0.023638336181640626, 0.02347225570678711, 0.023330816268920897, 0.023594047546386717, 0.0236759033203125, 0.023655616760253906, 0.023823040008544922, 0.023588544845581056, 0.023562496185302734, 0.02350707244873047, 0.023709375381469725, 0.02369977569580078, 0.023631872177124022, 0.02353561592102051, 0.02365023994445801, 0.023733407974243163, 0.023765920639038086, 0.023793664932250977, 0.023708896636962892, 0.02373446464538574, 0.023698015213012694, 0.023977088928222656, 0.023815040588378907, 0.023774368286132812, 0.023649120330810548, 0.023539968490600586, 0.02361523246765137, 0.023842239379882814, 0.023833152770996093, 0.02371583938598633, 0.02366624069213867, 0.023548351287841798, 0.023754751205444336, 0.0241112003326416, 0.024046592712402344, 0.023961727142333984, 0.023868192672729494, 0.023848447799682617, 0.023861408233642578, 0.02342092704772949, 0.023553632736206056, 0.023425024032592775, 0.023253408432006836, 0.023357343673706055, 0.023468128204345705, 0.02355766487121582, 0.023558271408081054, 0.023476224899291992, 0.02338377571105957, 0.023445888519287108, 0.02365670394897461, 0.023664192199707033, 0.023622079849243163, 0.02372198486328125, 0.023408639907836915, 0.023373823165893554, 0.02346188735961914, 0.023827999114990235, 0.023822816848754882, 0.023613248825073242, 0.02362553596496582, 0.02359129524230957, 0.023574560165405274, 0.02375052833557129, 0.023611488342285155, 0.023575616836547852, 0.023491519927978516, 0.023717695236206055, 0.024090816497802734, 0.023984128952026368, 0.023883520126342775, 0.02381987190246582, 0.02373699188232422, 0.023531007766723632, 0.023723615646362304, 0.023688095092773438, 0.023817920684814455, 0.02381376075744629, 0.023320768356323244, 0.023437183380126955, 0.023581024169921874, 0.02357872009277344, 0.023582912445068358, 0.023500799179077148, 0.023371776580810546, 0.023627616882324218, 0.023865631103515625, 0.023877504348754883, 0.023764640808105468, 0.023744863510131838, 0.023748607635498048, 0.023919904708862304, 0.02389852714538574, 0.023771455764770508, 0.023742464065551756, 0.02361248016357422, 0.023753664016723634, 0.024004512786865235, 0.024076383590698244, 0.023791616439819335, 0.023809951782226564]",tokens/s,42.26895357766238,, -4bit-gptq-exllama-v1-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neo,EleutherAI/gpt-neo-2.7B,EleutherAI/gpt-neo-2.7B,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neo,EleutherAI/gpt-neo-2.7B,EleutherAI/gpt-neo-2.7B,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -12290,7 +12290,7 @@ ChildProcessError: Traceback (most recent call last): ValueError: The model class you are passing has a `config_class` attribute that is not consistent with the config class you passed (model has and you passed . Fix one of those so they match! ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v1-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,deci,Deci/DeciLM-7B,Deci/DeciLM-7B,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,deci,Deci/DeciLM-7B,Deci/DeciLM-7B,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -12398,7 +12398,7 @@ ChildProcessError: Traceback (most recent call last): ValueError: RecurrentGemmaForCausalLM does not support an attention implementation through torch.nn.functional.scaled_dot_product_attention yet. Please request the support for this architecture: https://github.com/huggingface/transformers/issues/28005. If you believe this error is a bug, please open an issue in Transformers GitHub repository and load your model with the argument `attn_implementation=""eager""` meanwhile. Example: `model = AutoModel.from_pretrained(""openai/whisper-tiny"", attn_implementation=""eager"")` ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v1-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gptj,EleutherAI/gpt-j-6b,EleutherAI/gpt-j-6b,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gptj,EleutherAI/gpt-j-6b,EleutherAI/gpt-j-6b,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -12511,7 +12511,7 @@ ChildProcessError: Traceback (most recent call last): torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 68.00 MiB. GPU 0 has a total capacity of 14.74 GiB of which 46.12 MiB is free. Process 55922 has 14.69 GiB memory in use. Of the allocated memory 14.29 GiB is allocated by PyTorch, and 313.73 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v1-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neo,EleutherAI/gpt-neo-125m,EleutherAI/gpt-neo-125m,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neo,EleutherAI/gpt-neo-125m,EleutherAI/gpt-neo-125m,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -12644,7 +12644,7 @@ ChildProcessError: Traceback (most recent call last): AttributeError: 'EmissionsTracker' object has no attribute '_scheduler' ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v1-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neo,EleutherAI/gpt-neo-1.3B,EleutherAI/gpt-neo-1.3B,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neo,EleutherAI/gpt-neo-1.3B,EleutherAI/gpt-neo-1.3B,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -12857,7 +12857,7 @@ Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 4bit-gptq-exllama-v1-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-1.3b,EleutherAI/pythia-1.3b,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.223-212.873.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.4.0,,4.44.2,,0.34.0,,,,1.21.4,,,,0.12.0,,,MB,1584.623616,1596.915712,0.0,1201.668096,1189.151232,s,1,8.3438662109375,8.3438662109375,0.0,8.3438662109375,8.3438662109375,8.3438662109375,8.3438662109375,[8.3438662109375],,kWh,3.5593928066667024e-05,3.918479009886834e-06,1.0983897675998638e-05,5.04963047525525e-05,,MB,1721.929728,1794.048,0.0,1384.12032,1351.367168,s,10,0.41199757003784176,0.04119975700378418,0.00014483756846233586,0.041238176345825196,0.04129707107543945,0.04141133575439453,0.04150274749755859,"[0.04152560043334961, 0.041245952606201175, 0.04123302459716797, 0.04099168014526367, 0.041271678924560545, 0.04107120132446289, 0.04107059097290039, 0.04125964736938476, 0.041243328094482425, 0.04108486557006836]",tokens/s,6213.628880784091,kWh,1.24244321833335e-06,1.370178162340395e-07,8.233245357276679e-07,2.2027855702950574e-06,tokens/kWh,116216486.7303491,MB,1731.1744,1835.99104,0.0,1426.06336,1407.548416,s,10,10.576904418945313,1.0576904418945312,0.005352611792629986,1.0581462402343749,1.062176708984375,1.0641183227539062,1.0656716137695312,"[1.0660599365234376, 1.0612806396484376, 1.057637939453125, 1.04605126953125, 1.058654541015625, 1.055569580078125, 1.0571251220703124, 1.0617452392578124, 1.051779052734375, 1.0610010986328124]",tokens/s,59.5637414356838,kWh,3.0963543785417386e-05,3.414185807765365e-06,1.5384993986472118e-05,4.976272357965487e-05,tokens/kWh,1266007.8763405364,,s,630,10.574133043289184,0.016784338163951085,0.00033400692043555826,0.016730128288269044,0.017058774185180665,0.01727484941482544,0.01835901706695557,"[0.017328544616699217, 0.017398239135742188, 0.019746816635131836, 0.017338048934936522, 0.017328447341918945, 0.017270784378051757, 0.017137664794921875, 0.017006591796875, 0.017165567398071287, 0.017231744766235353, 0.017054559707641602, 0.018394336700439454, 0.01705196762084961, 0.016991968154907226, 0.016888799667358397, 0.01684432029724121, 0.016972192764282225, 0.016801664352416992, 0.016859136581420898, 0.016817760467529298, 0.01684115219116211, 0.01670550346374512, 0.016723968505859374, 0.01664169692993164, 0.016574560165405275, 0.016689407348632813, 0.016590368270874022, 0.016595424652099608, 0.016736255645751954, 0.016760576248168946, 0.016652671813964844, 0.01663167953491211, 0.01663801574707031, 0.01651456069946289, 0.016541280746459962, 0.016528480529785155, 0.016549055099487304, 0.016645952224731444, 0.016554752349853517, 0.01653555107116699, 0.016547840118408205, 0.016520736694335937, 0.016498687744140626, 0.01650326347351074, 0.016541696548461913, 0.01648227119445801, 0.016515104293823243, 0.016457696914672852, 0.016565439224243163, 0.018501760482788086, 0.01876652717590332, 0.016746623992919922, 0.016576416015625, 0.016617151260375978, 0.016764863967895508, 0.016619903564453125, 0.01661302375793457, 0.016699743270874024, 0.017123327255249024, 0.017155519485473635, 0.016878143310546875, 0.01692006492614746, 0.0168985595703125, 0.016728384017944336, 0.016622976303100587, 0.01651161575317383, 0.01637990379333496, 0.016433151245117187, 0.01636672019958496, 0.01643814468383789, 0.01662156867980957, 0.016702720642089844, 0.01674515151977539, 0.01691372871398926, 0.0167063045501709, 0.016736480712890626, 0.01679952049255371, 0.016730112075805666, 0.01668016052246094, 0.016747295379638674, 0.016592256546020506, 0.016581087112426757, 0.01652956771850586, 0.01678473663330078, 0.01681475257873535, 0.016930816650390625, 0.016777215957641603, 0.017083423614501953, 0.01678428840637207, 0.016756383895874024, 0.01665884780883789, 0.016875520706176757, 0.01662156867980957, 0.016691200256347655, 0.016614463806152342, 0.016645055770874023, 0.017143808364868163, 0.01717647933959961, 0.01663542366027832, 0.016659008026123048, 0.016740447998046876, 0.016750495910644533, 0.01670924758911133, 0.0169168643951416, 0.016764928817749023, 0.016730112075805666, 0.016912128448486326, 0.01686070442199707, 0.01667555236816406, 0.01684819221496582, 0.01668908882141113, 0.016730880737304686, 0.01666633605957031, 0.016752927780151368, 0.016631200790405275, 0.01797590446472168, 0.016836063385009765, 0.016726560592651367, 0.019449567794799803, 0.018272544860839842, 0.016942399978637696, 0.017070783615112304, 0.01716223907470703, 0.017334112167358397, 0.016857536315917968, 0.016805599212646485, 0.017108991622924806, 0.016888896942138673, 0.016927679061889647, 0.016903615951538085, 0.01703993606567383, 0.017079296112060546, 0.016873855590820313, 0.016760799407958986, 0.016554079055786132, 0.016622047424316406, 0.01680803108215332, 0.016699392318725585, 0.01674822425842285, 0.016685375213623045, 0.016904447555541994, 0.016625408172607423, 0.016751808166503908, 0.016812864303588866, 0.016680448532104493, 0.016653888702392577, 0.016587455749511718, 0.016648448944091798, 0.01679769515991211, 0.016662080764770507, 0.016693695068359375, 0.0166297607421875, 0.01663385581970215, 0.016570367813110352, 0.016699392318725585, 0.016598304748535155, 0.016769376754760743, 0.016673152923583984, 0.016644096374511717, 0.016539392471313478, 0.016615680694580078, 0.01653555107116699, 0.017320032119750976, 0.01705743980407715, 0.016765024185180662, 0.016692480087280272, 0.016739519119262695, 0.016801504135131835, 0.016760671615600586, 0.0166810245513916, 0.016773216247558592, 0.016748544692993163, 0.017016576766967773, 0.017344768524169923, 0.016951295852661134, 0.016928768157958983, 0.016832063674926758, 0.01681171226501465, 0.016934783935546874, 0.016954496383666993, 0.01690188789367676, 0.017088064193725584, 0.01713190460205078, 0.01683990478515625, 0.016741216659545897, 0.01664204788208008, 0.016516319274902345, 0.016431903839111327, 0.016578527450561525, 0.016645023345947266, 0.016582847595214844, 0.016561983108520507, 0.01684889602661133, 0.016512672424316407, 0.016559616088867187, 0.016452287673950194, 0.01645756721496582, 0.01655958366394043, 0.016626527786254883, 0.016451583862304688, 0.016530815124511718, 0.016783872604370118, 0.016490623474121092, 0.016556032180786134, 0.01645952033996582, 0.016435136795043947, 0.016539520263671875, 0.016550336837768555, 0.016581888198852538, 0.016547935485839844, 0.016520095825195313, 0.01647385597229004, 0.016523103713989257, 0.016455743789672853, 0.016530784606933593, 0.01644825553894043, 0.016472063064575194, 0.016457727432250976, 0.01657241630554199, 0.01646905517578125, 0.01728595161437988, 0.016627840042114258, 0.016590560913085937, 0.016470176696777344, 0.016766176223754883, 0.0165948486328125, 0.016559104919433593, 0.016447488784790038, 0.016590719223022462, 0.016558271408081054, 0.01665017509460449, 0.016562175750732423, 0.016570367813110352, 0.016554079055786132, 0.01651702308654785, 0.0164454402923584, 0.016453632354736326, 0.016493791580200194, 0.01653430366516113, 0.016523263931274415, 0.016559232711791994, 0.0165382080078125, 0.016419103622436523, 0.016434656143188477, 0.016575008392333983, 0.01659903907775879, 0.016850944519042968, 0.01692576026916504, 0.016927391052246093, 0.01716409683227539, 0.017289695739746095, 0.017037120819091797, 0.017079456329345703, 0.016754400253295897, 0.016631200790405275, 0.01657494354248047, 0.016573856353759766, 0.01667363166809082, 0.0166759033203125, 0.016702272415161132, 0.01671548843383789, 0.016795936584472655, 0.016744447708129884, 0.01671891212463379, 0.01672492790222168, 0.01665433692932129, 0.016670719146728515, 0.016861183166503906, 0.016855039596557618, 0.01683660888671875, 0.01676288032531738, 0.016762432098388673, 0.01680019187927246, 0.01683251190185547, 0.01688800048828125, 0.016942047119140626, 0.01675734329223633, 0.017033472061157226, 0.016856639862060548, 0.016818239212036134, 0.016914815902709962, 0.017098623275756834, 0.016900224685668944, 0.01684480094909668, 0.01678745651245117, 0.016787391662597656, 0.016863296508789063, 0.01679155158996582, 0.01699839973449707, 0.016766752243041992, 0.01661302375793457, 0.016728639602661133, 0.016898048400878905, 0.016931039810180664, 0.01682377624511719, 0.017117504119873048, 0.016855039596557618, 0.016854944229125975, 0.016803936004638673, 0.016902143478393555, 0.016943103790283204, 0.01705369567871094, 0.016817983627319337, 0.016791872024536133, 0.016781183242797853, 0.016744447708129884, 0.016693248748779296, 0.01676892852783203, 0.01670751953125, 0.01673846435546875, 0.01676697540283203, 0.016703487396240235, 0.016596607208251953, 0.016658176422119142, 0.01666227149963379, 0.01702230453491211, 0.0168799991607666, 0.01670172882080078, 0.01679555130004883, 0.016953407287597658, 0.016898080825805663, 0.016985952377319338, 0.016903392791748045, 0.016772031784057617, 0.016693248748779296, 0.01660927963256836, 0.0165928955078125, 0.016532512664794923, 0.016659423828125, 0.01661747169494629, 0.016900096893310547, 0.016664575576782227, 0.016639488220214844, 0.016574464797973632, 0.016961887359619142, 0.016873632431030273, 0.016631807327270508, 0.016727584838867188, 0.016798240661621094, 0.016813568115234375, 0.01678585624694824, 0.01679897689819336, 0.01671855926513672, 0.016750272750854493, 0.01678927993774414, 0.01684867286682129, 0.01686387252807617, 0.016807903289794923, 0.016863040924072266, 0.016690784454345704, 0.016626495361328125, 0.016631391525268553, 0.01653798484802246, 0.016498687744140626, 0.016488447189331054, 0.01646329689025879, 0.0164913272857666, 0.016444320678710937, 0.016633760452270507, 0.016689407348632813, 0.01655171203613281, 0.016547935485839844, 0.01662211227416992, 0.016566560745239257, 0.016674623489379883, 0.016979232788085937, 0.017308576583862305, 0.016637727737426757, 0.01656787109375, 0.016578943252563476, 0.016628000259399416, 0.016543807983398436, 0.016602880477905275, 0.01651321601867676, 0.016502784729003905, 0.016477888107299804, 0.01911020851135254, 0.017278175354003906, 0.016667776107788086, 0.016669536590576173, 0.01653334426879883, 0.01648044776916504, 0.016506879806518555, 0.016483999252319335, 0.0165382080078125, 0.016617216110229493, 0.016639999389648438, 0.01660723114013672, 0.01659903907775879, 0.016506879806518555, 0.01652115249633789, 0.01674025535583496, 0.01664620780944824, 0.01660323143005371, 0.01649446487426758, 0.016496768951416017, 0.01658233642578125, 0.016546112060546875, 0.01665023994445801, 0.016553216934204102, 0.016547584533691408, 0.016550943374633788, 0.01783724784851074, 0.017103616714477538, 0.0185262393951416, 0.016793600082397463, 0.016610912322998047, 0.016648607254028322, 0.01658060836791992, 0.016700576782226563, 0.016797664642333985, 0.016843648910522462, 0.01681203269958496, 0.016699392318725585, 0.016742399215698242, 0.016920576095581053, 0.01680384063720703, 0.01670966339111328, 0.016764543533325196, 0.016926752090454102, 0.017097024917602538, 0.017333343505859376, 0.016874399185180664, 0.01700864028930664, 0.016750175476074217, 0.01687183952331543, 0.016852447509765625, 0.01687606430053711, 0.01675596809387207, 0.01668118476867676, 0.016637760162353514, 0.016626399993896486, 0.016785408020019533, 0.017003807067871093, 0.01672643280029297, 0.01680732727050781, 0.016868255615234376, 0.016965375900268555, 0.016799936294555663, 0.016928831100463868, 0.01697996711730957, 0.016926752090454102, 0.016794143676757814, 0.01716806411743164, 0.017211103439331056, 0.017242303848266603, 0.01677676773071289, 0.01679657554626465, 0.01689948844909668, 0.01687196731567383, 0.016846847534179688, 0.016836063385009765, 0.016726335525512694, 0.016909664154052734, 0.016792512893676757, 0.016822208404541017, 0.016723775863647462, 0.016558271408081054, 0.016999679565429686, 0.01686809539794922, 0.01678745651245117, 0.01687481689453125, 0.016824287414550783, 0.01672435188293457, 0.01664601516723633, 0.016613439559936525, 0.016730144500732423, 0.01661929512023926, 0.016646751403808592, 0.016504480361938478, 0.01646134376525879, 0.01653446388244629, 0.01653708839416504, 0.016662912368774416, 0.016721920013427736, 0.016674783706665038, 0.016619455337524413, 0.016899391174316405, 0.0174021110534668, 0.016900640487670898, 0.017296672821044922, 0.017191648483276367, 0.01679769515991211, 0.016891904830932617, 0.01687731170654297, 0.016957696914672853, 0.016941055297851563, 0.016852991104125976, 0.016649375915527342, 0.016585151672363282, 0.016784000396728515, 0.016728992462158202, 0.0168723201751709, 0.017110015869140623, 0.017101472854614257, 0.016959840774536133, 0.01683456039428711, 0.016924671173095703, 0.017020256042480468, 0.01702355194091797, 0.016892000198364256, 0.016748544692993163, 0.01686028861999512, 0.017437568664550783, 0.01700454330444336, 0.016893888473510744, 0.01685264015197754, 0.01695568084716797, 0.01704560089111328, 0.016857120513916017, 0.016803647994995115, 0.016771263122558593, 0.016724319458007814, 0.01665603256225586, 0.016633216857910155, 0.016680864334106444, 0.017133440017700195, 0.016792352676391602, 0.016750656127929687, 0.01661747169494629, 0.016672767639160157, 0.01676643180847168, 0.016740896224975585, 0.016767999649047852, 0.016731136322021483, 0.017012224197387696, 0.01703558349609375, 0.01695120048522949, 0.017004352569580078, 0.017011167526245118, 0.01684480094909668, 0.01672208023071289, 0.016693088531494142, 0.016574304580688478, 0.01651113510131836, 0.016521472930908204, 0.016649984359741212, 0.016590368270874022, 0.01681046485900879, 0.01650908851623535, 0.016848735809326172, 0.016594944000244142, 0.016547840118408205, 0.016518592834472656, 0.0165545597076416, 0.016637535095214845, 0.016559871673583984, 0.017420448303222657, 0.016605247497558595, 0.016597087860107423, 0.016499040603637695, 0.01647542381286621, 0.01646870422363281, 0.016441343307495117, 0.016597024917602538, 0.01649660873413086, 0.016488544464111327, 0.016578079223632813, 0.016577024459838868, 0.01661939239501953, 0.01639833641052246, 0.0163918399810791, 0.016468320846557617, 0.01642905616760254, 0.016451040267944337, 0.016513792037963868, 0.016477983474731447, 0.016530048370361327, 0.016579999923706054, 0.01673916816711426, 0.016836639404296874, 0.016770687103271485, 0.016836896896362304, 0.016791072845458985, 0.016822816848754883, 0.01683977508544922, 0.01680476760864258, 0.016692672729492188, 0.016716352462768556, 0.016789535522460937, 0.016601055145263673, 0.01658470344543457, 0.016422592163085937, 0.016568639755249023, 0.01692451286315918, 0.016436672210693358, 0.017430944442749022, 0.0164150390625, 0.016357152938842774, 0.016462047576904296, 0.0166748161315918, 0.0167587833404541, 0.01715814399719238, 0.01707827186584473, 0.017126848220825195, 0.017044031143188476, 0.017405120849609376, 0.017578784942626952, 0.01694108772277832, 0.01699839973449707, 0.016914560317993165, 0.016905567169189454, 0.016779808044433592, 0.016989343643188475, 0.017310560226440428, 0.016952352523803713, 0.016819168090820312, 0.016731487274169923, 0.016722591400146484, 0.016719104766845704, 0.016609727859497072, 0.01663724708557129, 0.016616191864013672, 0.01661516761779785, 0.01656268882751465, 0.01658880043029785, 0.016766656875610353, 0.016654560089111328, 0.016663904190063476, 0.016601856231689454, 0.016633087158203125, 0.016589183807373047, 0.01680175971984863, 0.018142656326293947, 0.018082239151000976, 0.01688425636291504, 0.01669113540649414, 0.016951072692871095, 0.016978336334228517, 0.016882816314697267]",tokens/s,59.57935250302398,, 4bit-gptq-exllama-v1-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-1.4b,EleutherAI/pythia-1.4b,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.223-212.873.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.4.0,,4.44.2,,0.34.0,,,,1.21.4,,,,0.12.0,,,MB,1583.558656,1596.915712,0.0,1201.668096,1189.151232,s,1,8.3514072265625,8.3514072265625,0.0,8.3514072265625,8.3514072265625,8.3514072265625,8.3514072265625,[8.3514072265625],,kWh,3.551230208750174e-05,3.909955984242548e-06,1.0968342108001738e-05,5.039060017974603e-05,,MB,1587.183616,1794.048,0.0,1384.12032,1351.367168,s,10,0.41280178833007813,0.041280178833007816,0.0002324199835202943,0.04122892761230469,0.04140379981994628,0.04166441097259521,0.041872899894714356,"[0.04192502212524414, 0.04125302505493164, 0.04117756652832031, 0.04118697738647461, 0.04127635192871094, 0.041204830169677735, 0.0410423698425293, 0.041086944580078125, 0.04130281448364258, 0.04134588623046875]",tokens/s,6201.523521387975,kWh,1.2442479582092324e-06,1.372182518467848e-07,8.22242974578723e-07,2.20370918463474e-06,tokens/kWh,116167778.30076134,MB,1592.111104,1835.99104,0.0,1426.06336,1407.548416,s,10,10.592636230468752,1.0592636230468753,0.004145041345814589,1.0582247924804689,1.0626305419921875,1.0657195678710938,1.0681907885742188,"[1.0590074462890624, 1.05448046875, 1.061748046875, 1.06880859375, 1.0547740478515626, 1.057442138671875, 1.061944091796875, 1.0614659423828126, 1.057037109375, 1.0559283447265626]",tokens/s,59.47527945761627,kWh,3.0436203563040093e-05,3.3566527589631704e-06,1.536185330602108e-05,4.915470962802435e-05,tokens/kWh,1281667.6260881033,,s,630,10.589003307342528,0.016807941757686556,0.00032476302278186965,0.016749680519104006,0.016978409004211426,0.017119202995300294,0.017825050716400153,"[0.017150848388671876, 0.017287168502807617, 0.016746496200561522, 0.016875520706176757, 0.01684217643737793, 0.016746112823486328, 0.01671673583984375, 0.01660313606262207, 0.016697216033935546, 0.01676710319519043, 0.016852991104125976, 0.01664963150024414, 0.016642656326293945, 0.016740352630615234, 0.01662761688232422, 0.01687126350402832, 0.016965856552124025, 0.016768224716186525, 0.0168437442779541, 0.01672707176208496, 0.01672025680541992, 0.016713823318481445, 0.01672540855407715, 0.01683888053894043, 0.01675132751464844, 0.016748544692993163, 0.016778783798217775, 0.016869855880737306, 0.016906240463256835, 0.016884895324707033, 0.016795967102050783, 0.016860864639282228, 0.01676153564453125, 0.016793760299682617, 0.01678745651245117, 0.016789344787597655, 0.017066144943237306, 0.017964927673339844, 0.016765056610107423, 0.01677644729614258, 0.01664896011352539, 0.016655359268188476, 0.0167490234375, 0.016868896484375, 0.016741439819335936, 0.016914304733276368, 0.01686124801635742, 0.017123327255249024, 0.016726015090942382, 0.016764928817749023, 0.016697343826293946, 0.016793535232543944, 0.01669536018371582, 0.016729663848876954, 0.016681407928466795, 0.016717119216918944, 0.016652095794677736, 0.016714624404907227, 0.01670528030395508, 0.01675699234008789, 0.01669468879699707, 0.016810527801513674, 0.016659616470336914, 0.016749183654785157, 0.016668607711791992, 0.01669945526123047, 0.016674144744873047, 0.016882335662841797, 0.016611072540283205, 0.016762752532958985, 0.016673152923583984, 0.016633024215698244, 0.016607072830200194, 0.016788448333740234, 0.01660054397583008, 0.016798112869262697, 0.016689279556274413, 0.016690431594848634, 0.0166878719329834, 0.016842752456665038, 0.016728063583374024, 0.01683660888671875, 0.016707231521606445, 0.016718175888061522, 0.0168787841796875, 0.01697260856628418, 0.016783327102661134, 0.01677110481262207, 0.016748544692993163, 0.0167956485748291, 0.016639999389648438, 0.01669526481628418, 0.01668239974975586, 0.016749055862426757, 0.01680499267578125, 0.017029375076293946, 0.017089279174804687, 0.01679270362854004, 0.016719968795776367, 0.01685772705078125, 0.016692703247070312, 0.016610208511352538, 0.016727712631225584, 0.0165449275970459, 0.01688675117492676, 0.016852319717407225, 0.016724544525146483, 0.016691295623779297, 0.0166748161315918, 0.016713727951049806, 0.01661952018737793, 0.016750591278076172, 0.01667036819458008, 0.016617183685302735, 0.016656511306762694, 0.016629535675048827, 0.01650524711608887, 0.01678995132446289, 0.016786367416381835, 0.016861568450927733, 0.016788127899169922, 0.016697248458862304, 0.016739519119262695, 0.016595775604248048, 0.016594911575317384, 0.016773151397705077, 0.01666454315185547, 0.016793407440185548, 0.01705923271179199, 0.016935712814331056, 0.01679769515991211, 0.01662512016296387, 0.01689244842529297, 0.016794784545898438, 0.016609119415283202, 0.016607967376708985, 0.016785696029663087, 0.016838623046875, 0.016666591644287108, 0.01674799919128418, 0.01674710464477539, 0.016951295852661134, 0.017088512420654296, 0.01672563171386719, 0.016713216781616212, 0.01674025535583496, 0.016724960327148437, 0.01664780807495117, 0.01675712013244629, 0.016729663848876954, 0.01662816047668457, 0.016697343826293946, 0.016961536407470702, 0.016728063583374024, 0.016752639770507814, 0.01668057632446289, 0.016719968795776367, 0.016828447341918944, 0.01676860809326172, 0.016793695449829102, 0.016886335372924804, 0.016971967697143556, 0.017194911956787108, 0.01717033576965332, 0.016977088928222656, 0.01704140853881836, 0.0169268798828125, 0.016884479522705078, 0.016854944229125975, 0.016809120178222656, 0.016867647171020506, 0.016760671615600586, 0.01700524711608887, 0.017864704132080078, 0.017680383682250975, 0.017104032516479493, 0.017005407333374023, 0.01711916732788086, 0.01694918441772461, 0.01677939224243164, 0.016744447708129884, 0.01671507263183594, 0.01667910385131836, 0.01684940719604492, 0.016678911209106445, 0.01672985649108887, 0.01665795135498047, 0.016636640548706054, 0.016772224426269532, 0.016833536148071288, 0.016675840377807616, 0.016683008193969725, 0.016719871520996094, 0.01668819236755371, 0.016677824020385743, 0.016683008193969725, 0.016769023895263673, 0.016721920013427736, 0.016963584899902344, 0.017119232177734374, 0.016730112075805666, 0.01660416030883789, 0.0168089599609375, 0.016657888412475588, 0.016544288635253906, 0.0166014404296875, 0.016544511795043945, 0.01658358383178711, 0.016699199676513673, 0.01747577667236328, 0.020911552429199218, 0.021582239151000975, 0.01703651237487793, 0.016823232650756834, 0.016908031463623047, 0.016689407348632813, 0.016858272552490235, 0.016670751571655273, 0.016933439254760742, 0.016679168701171875, 0.016742399215698242, 0.016693248748779296, 0.01678745651245117, 0.016702592849731444, 0.0171276798248291, 0.0171812801361084, 0.016777023315429688, 0.0167174072265625, 0.016748512268066406, 0.016695968627929686, 0.016877407073974608, 0.016871583938598632, 0.01666864013671875, 0.01665363121032715, 0.016890592575073242, 0.016731584548950195, 0.01673263931274414, 0.016752735137939453, 0.016708992004394532, 0.01670412826538086, 0.016852991104125976, 0.016746496200561522, 0.01667024040222168, 0.016656448364257812, 0.016756160736083985, 0.016673215866088865, 0.016828128814697266, 0.01674671936035156, 0.01690809631347656, 0.016742687225341796, 0.0191779842376709, 0.01718003273010254, 0.01734294319152832, 0.016876575469970703, 0.01684124755859375, 0.016887231826782225, 0.016709856033325195, 0.01688991928100586, 0.01694371223449707, 0.016648191452026367, 0.016611263275146483, 0.016643295288085936, 0.016622432708740233, 0.01663795280456543, 0.01666662406921387, 0.01683865547180176, 0.016602655410766602, 0.016743104934692384, 0.016647680282592774, 0.016760671615600586, 0.016683584213256837, 0.01685286331176758, 0.016611551284790037, 0.016763776779174806, 0.016698432922363282, 0.01693270492553711, 0.016570367813110352, 0.016774560928344725, 0.016681568145751953, 0.01665843200683594, 0.016583744049072265, 0.016813119888305663, 0.01696460723876953, 0.016740447998046876, 0.01677187156677246, 0.016670719146728515, 0.016655359268188476, 0.01660211181640625, 0.016645824432373047, 0.01678163146972656, 0.016603328704833983, 0.01672502326965332, 0.016577184677124022, 0.01676915168762207, 0.016704704284667967, 0.01694921684265137, 0.016802656173706056, 0.016879039764404295, 0.016646080017089844, 0.016617311477661132, 0.016681695938110353, 0.016738367080688477, 0.01660678482055664, 0.016839040756225585, 0.01662099266052246, 0.01686300849914551, 0.01670639991760254, 0.01682841682434082, 0.016660480499267577, 0.016683008193969725, 0.016717824935913086, 0.01679120063781738, 0.016662879943847655, 0.01677107238769531, 0.016732160568237304, 0.01710095977783203, 0.017058431625366213, 0.016781312942504883, 0.016774431228637695, 0.01671446418762207, 0.01663702392578125, 0.016760992050170898, 0.016763647079467772, 0.017102848052978514, 0.016842752456665038, 0.01677836799621582, 0.0168143367767334, 0.01681878471374512, 0.016817535400390625, 0.016865951538085937, 0.01686499214172363, 0.016711519241333007, 0.017183168411254883, 0.016670719146728515, 0.01690323257446289, 0.016747264862060546, 0.016810176849365234, 0.01676803207397461, 0.016712671279907228, 0.01697996711730957, 0.016957536697387695, 0.016762176513671876, 0.01680182456970215, 0.016696895599365234, 0.016747520446777343, 0.016631807327270508, 0.016710975646972655, 0.01673664093017578, 0.016744768142700196, 0.016693248748779296, 0.016648191452026367, 0.016622751235961915, 0.01667568016052246, 0.016692703247070312, 0.016910879135131836, 0.01663795280456543, 0.016627328872680664, 0.016654720306396486, 0.01676892852783203, 0.016601184844970703, 0.016717824935913086, 0.01657241630554199, 0.016764928817749023, 0.016726015090942382, 0.01682636833190918, 0.01667430305480957, 0.01674006462097168, 0.01675254440307617, 0.01664899253845215, 0.016706911087036133, 0.01673664093017578, 0.016615680694580078, 0.01696780776977539, 0.016760639190673828, 0.01669548797607422, 0.016828479766845702, 0.017229759216308593, 0.016897247314453124, 0.01677587127685547, 0.016596960067749023, 0.016689344406127928, 0.01670515251159668, 0.016693695068359375, 0.016680831909179687, 0.01669340705871582, 0.01664614486694336, 0.016777215957641603, 0.01682841682434082, 0.016846847534179688, 0.016632863998413086, 0.016759456634521483, 0.016813535690307618, 0.016672767639160157, 0.016622432708740233, 0.01671548843383789, 0.01761065673828125, 0.016715263366699217, 0.016663103103637694, 0.0168656005859375, 0.0167521915435791, 0.016884159088134766, 0.016635967254638673, 0.01665836715698242, 0.016793600082397463, 0.016844287872314453, 0.01672038459777832, 0.016852991104125976, 0.016737823486328126, 0.01661952018737793, 0.01662828826904297, 0.016746463775634764, 0.01668835258483887, 0.017320415496826173, 0.016793855667114256, 0.016687103271484375, 0.01721958351135254, 0.01683443260192871, 0.0166843204498291, 0.016620384216308594, 0.01677516746520996, 0.016807903289794923, 0.016643392562866212, 0.016706079483032227, 0.016982175827026366, 0.016803871154785155, 0.016752639770507814, 0.016772928237915038, 0.016718015670776368, 0.0168571834564209, 0.017033119201660157, 0.017881088256835938, 0.018191904067993165, 0.0169783992767334, 0.01695692825317383, 0.016949535369873047, 0.016873695373535155, 0.016829504013061523, 0.016686016082763672, 0.01683033561706543, 0.017127552032470704, 0.017394943237304686, 0.017176448822021486, 0.017006528854370116, 0.01684294319152832, 0.01690006446838379, 0.01672400093078613, 0.0169998722076416, 0.016795423507690428, 0.016676927566528322, 0.016777952194213866, 0.016774784088134764, 0.01687923240661621, 0.016631776809692383, 0.017727968215942382, 0.01667308807373047, 0.01681203269958496, 0.017625087738037108, 0.01705366325378418, 0.016821823120117187, 0.016922592163085937, 0.016652351379394532, 0.01666502380371094, 0.016693248748779296, 0.016700864791870117, 0.016702016830444335, 0.01684889602661133, 0.017160192489624023, 0.01699190330505371, 0.016968032836914063, 0.017003711700439454, 0.016975744247436524, 0.017044416427612306, 0.017059743881225584, 0.016730207443237305, 0.01676233673095703, 0.016636447906494142, 0.016961088180541994, 0.016913888931274414, 0.016587232589721678, 0.016717727661132813, 0.016880224227905274, 0.01683456039428711, 0.01679974365234375, 0.016664575576782227, 0.016773311614990235, 0.01680899238586426, 0.016783456802368164, 0.016667327880859374, 0.016664575576782227, 0.01677926445007324, 0.016685056686401366, 0.016662687301635743, 0.016770591735839845, 0.01677343940734863, 0.016828191757202147, 0.016908607482910155, 0.017033119201660157, 0.016707584381103514, 0.016715776443481444, 0.016714048385620118, 0.01665158462524414, 0.01660713577270508, 0.016664768218994142, 0.017252191543579102, 0.01720966339111328, 0.016949247360229493, 0.01682841682434082, 0.016717824935913086, 0.01666819190979004, 0.016654367446899413, 0.016667072296142577, 0.016772159576416014, 0.01681056022644043, 0.016847232818603515, 0.016743488311767578, 0.016679040908813475, 0.017023807525634767, 0.01680588722229004, 0.016726015090942382, 0.01677107238769531, 0.016639135360717774, 0.016784223556518554, 0.016730112075805666, 0.01674425506591797, 0.016766176223754883, 0.01685321617126465, 0.01687424087524414, 0.01676288032531738, 0.016732160568237304, 0.0167956485748291, 0.01671513557434082, 0.016804479598999025, 0.016713247299194336, 0.01679408073425293, 0.016701440811157226, 0.016728063583374024, 0.016709632873535156, 0.01677107238769531, 0.01660723114013672, 0.016742399215698242, 0.016676864624023437, 0.016877567291259766, 0.016859136581420898, 0.016898048400878905, 0.016685056686401366, 0.01665433692932129, 0.01675382423400879, 0.016724191665649413, 0.01686140823364258, 0.016789920806884767, 0.016644096374511717, 0.016770143508911133, 0.016892831802368165, 0.01674985694885254, 0.017081056594848633, 0.01682431983947754, 0.01664614486694336, 0.016633216857910155, 0.01665292739868164, 0.01663385581970215, 0.016573728561401366, 0.01682896041870117, 0.01662585639953613, 0.0169716796875, 0.016633567810058595, 0.016820512771606445, 0.01674665641784668, 0.016906240463256835, 0.016701440811157226, 0.01684000015258789, 0.016886463165283205, 0.01701478385925293, 0.016760831832885743, 0.016752639770507814, 0.016764928817749023, 0.017043455123901367, 0.016883712768554687, 0.01684480094909668, 0.016752511978149413, 0.01679372787475586, 0.01666646385192871, 0.016623008728027345, 0.01678976058959961, 0.016808448791503908, 0.01679769515991211, 0.016678911209106445, 0.01667635154724121, 0.0170644474029541, 0.016901567459106447, 0.01697849655151367, 0.016856992721557617, 0.01664771270751953, 0.016756767272949218, 0.01664259147644043, 0.016928096771240235, 0.01677788734436035, 0.01668611145019531, 0.016563167572021486, 0.016688671112060547, 0.016616992950439453, 0.01663702392578125, 0.01674950408935547, 0.016690080642700195, 0.016643552780151366, 0.01669174385070801, 0.01666662406921387, 0.01667433547973633, 0.016650880813598633, 0.01671561622619629, 0.016606239318847655, 0.016683584213256837, 0.016916223526000976, 0.01692972755432129, 0.016776512145996094, 0.01671824073791504, 0.01665433692932129, 0.016695295333862305, 0.016760831832885743, 0.016801599502563477, 0.016659744262695314, 0.016705535888671876, 0.01676585578918457, 0.016900096893310547, 0.016668031692504883, 0.016681215286254884, 0.016628095626831055, 0.01664204788208008, 0.01660259246826172, 0.016660192489624023, 0.016614208221435545]",tokens/s,59.49568450537277,, -4bit-gptq-exllama-v1-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -12902,7 +12902,7 @@ ChildProcessError: Traceback (most recent call last): self._buffers[key] = fn(buf) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1160, in convert return t.to( -torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 70.00 MiB. GPU 0 has a total capacity of 14.74 GiB of which 16.12 MiB is free. Process 20925 has 14.72 GiB memory in use. Of the allocated memory 14.44 GiB is allocated by PyTorch, and 187.67 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 70.00 MiB. GPU 0 has a total capacity of 14.74 GiB of which 16.12 MiB is free. Process 20887 has 14.72 GiB memory in use. Of the allocated memory 14.44 GiB is allocated by PyTorch, and 187.67 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 4bit-gptq-exllama-v1-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-12b,EleutherAI/pythia-12b,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.223-212.873.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.4.0,,4.44.2,,0.34.0,,,,1.21.4,,,,0.12.0,,,MB,6765.789184,7762.542592,0.0,7367.294976,7351.94368,s,1,12.5058095703125,12.5058095703125,0.0,12.5058095703125,12.5058095703125,12.5058095703125,12.5058095703125,[12.5058095703125],,kWh,0.00015767173855417126,1.738523505041432e-05,5.054559599200134e-05,0.00022560256959658692,,MB,1635.319808,8404.271104,0.0,7994.343424,7863.794176,s,10,2.9980195617675776,0.29980195617675787,0.0004853329511504965,0.2998737182617187,0.3003250793457031,0.3003877288818359,0.3004378485107422,"[0.2996583557128906, 0.30008908081054686, 0.29907073974609377, 0.29927716064453125, 0.29923199462890626, 0.30045037841796873, 0.30015191650390627, 0.30027667236328126, 0.3003111572265625, 0.2995021057128906]",tokens/s,853.8970301083259,kWh,8.76525808872551e-06,9.666528461919464e-07,5.797406598705881e-06,1.5529317533623337e-05,tokens/kWh,16484948.514042618,MB,1641.824256,8676.900864,0.0,8266.973184,8120.408064,s,10,24.660447021484373,2.4660447021484373,0.0022546624102290007,2.4662266845703122,2.4684953125,2.4685719970703124,2.4686333447265625,"[2.465872314453125, 2.465596923828125, 2.4603427734375, 2.46468408203125, 2.467619873046875, 2.46687158203125, 2.468648681640625, 2.46575146484375, 2.4665810546875, 2.468478271484375]",tokens/s,25.546982155316936,kWh,7.217162492544173e-05,7.960560736861764e-06,4.812663654049395e-05,0.00012825882220279745,tokens/kWh,491194.2813601317,,s,630,24.654574352264408,0.03913424500359429,0.0003454978796092168,0.0391464958190918,0.03957495574951172,0.03966784496307373,0.03983591388702393,"[0.03908633422851562, 0.03847494506835938, 0.03833103942871094, 0.03829171371459961, 0.038526878356933594, 0.03899097442626953, 0.03889823913574219, 0.03885702514648438, 0.038631393432617185, 0.03855363082885742, 0.03917004776000976, 0.03863142395019531, 0.03844710540771484, 0.0385986557006836, 0.03871539306640625, 0.038866943359375, 0.03882393646240234, 0.038727680206298826, 0.03872972869873047, 0.03907379150390625, 0.03919772720336914, 0.03914236831665039, 0.03912483215332031, 0.03905142211914062, 0.03930521774291992, 0.039206912994384766, 0.039188255310058595, 0.03881596755981445, 0.038830078125, 0.0387665901184082, 0.039073055267333984, 0.03901513671875, 0.03908943939208984, 0.03895574569702148, 0.039374526977539064, 0.03924166488647461, 0.03908415985107422, 0.03921529769897461, 0.03938924789428711, 0.03960627365112305, 0.03947315216064453, 0.03955055999755859, 0.039526622772216795, 0.03944451141357422, 0.03933552169799805, 0.03915568161010742, 0.03917168045043945, 0.03952742385864258, 0.039373855590820316, 0.03941020965576172, 0.039637439727783205, 0.03946495819091797, 0.039247966766357424, 0.03930908966064453, 0.03944979095458984, 0.03929980850219727, 0.03959215927124023, 0.03966918563842774, 0.03949539184570312, 0.039905982971191405, 0.039701793670654295, 0.03977865600585938, 0.03989763259887695, 0.03919830322265625, 0.03895702362060547, 0.038859455108642575, 0.03898137664794922, 0.03891225433349609, 0.03890291213989258, 0.038277568817138674, 0.03859632110595703, 0.03843468856811524, 0.03857705688476563, 0.03897951889038086, 0.03896054458618164, 0.038754913330078126, 0.03876249694824219, 0.03929644775390625, 0.039166526794433595, 0.03902668762207031, 0.038771743774414065, 0.03885564804077148, 0.038776832580566405, 0.038809600830078124, 0.038819839477539066, 0.03923747253417969, 0.039166175842285156, 0.03884806442260742, 0.03942643356323242, 0.03916156768798828, 0.03917238235473633, 0.03909836959838867, 0.03903039932250976, 0.039541118621826174, 0.03944160079956055, 0.03918316650390625, 0.03896319961547851, 0.0388218879699707, 0.03886899185180664, 0.03897100830078125, 0.039123329162597656, 0.039251518249511716, 0.039039615631103516, 0.03924332809448242, 0.039198974609375, 0.03910860824584961, 0.03946086502075195, 0.03938230514526367, 0.039402206420898436, 0.0392540168762207, 0.03923763275146484, 0.03910806274414062, 0.03962246322631836, 0.039778144836425784, 0.03965017700195313, 0.03947724914550781, 0.03935635375976562, 0.03935785675048828, 0.039365280151367185, 0.03947724914550781, 0.039370750427246096, 0.03933184051513672, 0.0393072624206543, 0.03922739028930664, 0.03951113510131836, 0.039443359375, 0.03890985488891602, 0.03852912139892578, 0.038545406341552735, 0.03855974578857422, 0.0385654411315918, 0.038521278381347654, 0.0384224967956543, 0.03846352005004883, 0.03845657730102539, 0.03877964782714844, 0.03866419219970703, 0.0385629768371582, 0.03856083297729492, 0.039030017852783205, 0.038930976867675784, 0.03887891387939453, 0.038844734191894534, 0.03880755233764648, 0.03870515060424805, 0.038578174591064454, 0.03853107070922852, 0.0388955192565918, 0.03902268981933594, 0.038834175109863284, 0.03908748626708984, 0.03920355224609375, 0.03910851287841797, 0.038946495056152344, 0.038963264465332034, 0.0388139533996582, 0.03936793518066406, 0.039242496490478514, 0.03906755065917969, 0.03894895935058594, 0.03896934509277344, 0.03911270523071289, 0.03892784118652344, 0.03896368026733398, 0.03908169555664062, 0.03919625473022461, 0.03929065704345703, 0.039297855377197266, 0.03925609588623047, 0.039250049591064456, 0.039257953643798825, 0.03962860870361328, 0.03947555160522461, 0.03938508987426758, 0.039413761138916016, 0.039452671051025394, 0.0394947509765625, 0.03954166412353516, 0.03939123153686523, 0.03936880111694336, 0.039245246887207035, 0.039252574920654294, 0.039474880218505856, 0.03960639953613281, 0.039446590423583984, 0.03951424026489258, 0.0395359992980957, 0.03946547317504883, 0.03937212753295898, 0.03901440048217773, 0.038747135162353515, 0.038492225646972654, 0.03894524765014649, 0.039061920166015625, 0.03886678314208984, 0.03855791854858399, 0.0385167350769043, 0.038553600311279294, 0.03877068710327149, 0.03877273559570313, 0.03873382568359375, 0.038836223602294925, 0.03881951904296875, 0.03884054565429688, 0.038757823944091795, 0.03915433502197266, 0.03899135971069336, 0.038992385864257816, 0.03890716934204102, 0.03875094223022461, 0.03882393646240234, 0.03887513732910156, 0.03874540710449219, 0.038957439422607425, 0.03901472091674805, 0.03894831848144531, 0.03887760162353516, 0.03935801696777344, 0.03922371292114258, 0.03917536163330078, 0.039203422546386715, 0.03935475158691406, 0.03919257736206055, 0.03926012802124024, 0.039423103332519534, 0.039265182495117186, 0.03904307174682617, 0.03892019271850586, 0.03905036926269531, 0.039120960235595706, 0.039050048828125, 0.03911385726928711, 0.03934067153930664, 0.039272705078125, 0.039284320831298826, 0.0395145263671875, 0.03934400177001953, 0.03943027114868164, 0.039460159301757815, 0.03928543853759765, 0.03971859359741211, 0.03963935852050781, 0.039413761138916016, 0.03942102432250977, 0.039424671173095706, 0.039551231384277345, 0.039624385833740235, 0.039548961639404294, 0.03943164825439453, 0.03959276962280273, 0.0395142707824707, 0.03946684646606445, 0.03901033782958984, 0.03861577606201172, 0.03848732757568359, 0.038486560821533206, 0.038798656463623044, 0.03878579330444336, 0.03866195297241211, 0.0386761589050293, 0.038765182495117186, 0.03880956649780273, 0.03876457595825195, 0.03870684814453125, 0.03892870330810547, 0.038895294189453124, 0.03881350326538086, 0.03877737426757812, 0.038959102630615236, 0.038958942413330075, 0.03873603057861328, 0.03901792144775391, 0.03902521514892578, 0.0393994255065918, 0.039016448974609375, 0.039124385833740234, 0.03902912139892578, 0.0390043830871582, 0.038735870361328126, 0.038856254577636716, 0.03909062576293945, 0.03914956665039063, 0.03905945587158203, 0.0391734733581543, 0.03916422271728515, 0.039063583374023436, 0.039512382507324216, 0.03952640151977539, 0.03931545639038086, 0.0393359375, 0.03922518539428711, 0.03938524627685547, 0.03928617477416992, 0.03904982376098633, 0.03920896148681641, 0.03936275100708008, 0.03937849426269531, 0.03938451385498047, 0.039405536651611325, 0.039410526275634766, 0.03946022415161133, 0.03937139129638672, 0.039684097290039064, 0.03947520065307617, 0.03959807968139648, 0.03953664016723633, 0.03949158477783203, 0.03956252670288086, 0.03959676742553711, 0.03955279922485352, 0.03957753753662109, 0.0394666862487793, 0.03974614334106445, 0.03972463989257812, 0.0400654411315918, 0.03949135971069336, 0.03877068710327149, 0.03853952026367188, 0.03837952041625976, 0.03858790588378906, 0.03865983963012695, 0.03862579345703125, 0.038604225158691406, 0.038472511291503905, 0.03874816131591797, 0.03878448104858399, 0.03890230560302734, 0.03886489486694336, 0.03895296096801758, 0.03865727996826172, 0.03863808059692383, 0.038965503692626954, 0.03907583999633789, 0.038873153686523436, 0.03879740905761719, 0.03873484802246094, 0.039008960723876954, 0.03924595260620117, 0.03937283325195313, 0.039231487274169925, 0.03914342498779297, 0.0390041618347168, 0.03895500946044922, 0.0393809928894043, 0.03894076919555664, 0.03896105575561523, 0.03904512023925781, 0.03896115112304688, 0.03902668762207031, 0.03917824172973633, 0.039153568267822264, 0.03915935897827148, 0.0393240966796875, 0.039346271514892575, 0.039327743530273435, 0.039362560272216796, 0.039589088439941404, 0.039658432006835935, 0.03944755172729492, 0.03941462326049805, 0.039399265289306644, 0.039349536895751956, 0.03926713562011719, 0.03971263885498047, 0.03972713470458984, 0.039562751770019534, 0.039737888336181644, 0.0395470085144043, 0.039346176147460936, 0.03926425552368164, 0.03931545639038086, 0.039329792022705076, 0.03933184051513672, 0.03946086502075195, 0.039550975799560545, 0.039637054443359375, 0.0394936637878418, 0.03943734359741211, 0.038967262268066405, 0.03844409561157226, 0.038675422668457034, 0.038801406860351564, 0.03873750305175781, 0.03901603317260742, 0.03892102432250977, 0.03878908920288086, 0.03875433731079102, 0.038596607208251955, 0.038604286193847655, 0.038631519317626956, 0.03864617538452148, 0.0388485107421875, 0.038917537689208984, 0.038741695404052735, 0.03872604751586914, 0.039135295867919924, 0.039258560180664065, 0.03897923278808594, 0.039035232543945315, 0.03896319961547851, 0.03898303985595703, 0.038824577331542966, 0.03889561462402344, 0.038828033447265625, 0.0389939193725586, 0.03902054214477539, 0.03918950271606445, 0.03943936157226562, 0.03926835250854492, 0.03915161514282227, 0.039210945129394534, 0.03958585739135742, 0.03938508987426758, 0.03918438339233398, 0.03915375900268555, 0.03917404937744141, 0.03946086502075195, 0.039241695404052736, 0.03900617599487305, 0.039110721588134764, 0.03909836959838867, 0.0390978889465332, 0.03927702331542969, 0.039354366302490236, 0.03925299072265625, 0.03979270553588867, 0.03973388671875, 0.03953190231323242, 0.039494590759277345, 0.03949552154541015, 0.03936819076538086, 0.03980579376220703, 0.039804737091064454, 0.039653377532958986, 0.03972259140014649, 0.03957564926147461, 0.03978604888916016, 0.03972787094116211, 0.03974092864990234, 0.03990784072875977, 0.0398287353515625, 0.038836223602294925, 0.03855769729614258, 0.03857920074462891, 0.03857926559448242, 0.03845929718017578, 0.03849628829956055, 0.038991870880126955, 0.03892633438110352, 0.038866943359375, 0.03908169555664062, 0.03920515060424805, 0.039008255004882815, 0.038729217529296874, 0.03857049560546875, 0.03867766571044922, 0.03879100799560547, 0.038832897186279296, 0.038840576171875, 0.03879100799560547, 0.03859881591796875, 0.03896115112304688, 0.038948863983154294, 0.038868896484375, 0.03919267272949219, 0.03923763275146484, 0.03917004776000976, 0.038967201232910156, 0.039038143157958984, 0.03898780822753906, 0.03888217544555664, 0.03883318328857422, 0.03905436706542969, 0.03907372665405273, 0.03896310424804687, 0.03932579040527344, 0.0394420166015625, 0.03948118209838867, 0.03928326416015625, 0.039316608428955076, 0.03961494445800781, 0.039597793579101564, 0.039527103424072264, 0.03927449417114258, 0.03925596618652344, 0.039209056854248046, 0.03916377639770508, 0.03912307357788086, 0.039180030822753904, 0.03932928085327148, 0.03935308837890625, 0.03930252838134766, 0.03923212814331055, 0.039378944396972655, 0.03937279891967774, 0.03932947158813477, 0.039522624969482424, 0.03953251266479492, 0.039507999420166015, 0.03958736038208008, 0.03966620635986328, 0.039851966857910155, 0.03983769607543945, 0.03962265777587891, 0.038776832580566405, 0.03861276626586914, 0.03865827178955078, 0.0387209587097168, 0.038591041564941406, 0.03849324798583984, 0.03844944000244141, 0.03865628814697265, 0.038711681365966796, 0.038587455749511716, 0.03879199981689453, 0.038801345825195316, 0.03873196792602539, 0.03877273559570313, 0.039106559753417966, 0.03906150436401367, 0.039016448974609375, 0.03899347305297852, 0.039239776611328124, 0.03908339309692383, 0.0388271369934082, 0.03894051361083985, 0.03888329696655273, 0.03881129455566406, 0.039161407470703125, 0.03909846496582031, 0.03899260711669922, 0.038997791290283204, 0.038964542388916015, 0.03905014419555664, 0.03933996963500976, 0.03930527877807617, 0.03916377639770508, 0.03909235382080078, 0.03899369430541992, 0.03901871871948242, 0.03913900756835938, 0.039113025665283206, 0.03903398513793945, 0.03906233596801758, 0.039465023040771485, 0.03954483032226563, 0.03953987121582031, 0.0396317138671875, 0.03934793472290039, 0.039366943359375, 0.03929817581176758, 0.039433216094970705, 0.039363872528076174, 0.03964364624023437, 0.03957487869262695, 0.03942211151123047, 0.039420127868652344, 0.03935171127319336, 0.03979702377319336, 0.03962265777587891, 0.0396192626953125, 0.03957350540161133, 0.03949702453613281, 0.03953529739379883, 0.03953664016723633, 0.039593982696533206, 0.03954867172241211, 0.03917619323730469, 0.03869696044921875, 0.03838771057128906, 0.03904307174682617, 0.03878911972045898, 0.038899711608886715, 0.03872870254516601, 0.03873689651489258, 0.038735870361328126, 0.03878815841674805, 0.03876927947998047, 0.03884268951416016, 0.03889299011230469, 0.03895552062988281, 0.03900831985473633, 0.03900131225585937, 0.03912783813476563, 0.03897139358520508, 0.038860801696777345, 0.038793182373046874, 0.038846752166748044, 0.039161441802978515, 0.03918044662475586, 0.03889766311645508, 0.038876480102539065, 0.038769344329833984, 0.03886297607421875, 0.03880467224121094, 0.03898643112182617, 0.0389956169128418, 0.039036991119384766, 0.038803775787353514, 0.03935990524291992, 0.03929964828491211, 0.03949977493286133, 0.03919647979736328, 0.039184574127197266, 0.03951395034790039, 0.039591712951660155, 0.03950758361816406, 0.039524513244628905, 0.03971535873413086, 0.03946092987060547, 0.03930918502807617, 0.03919257736206055, 0.03938111877441406, 0.03942399978637695, 0.039376609802246096, 0.039385215759277344, 0.039333343505859375, 0.03920966339111328, 0.03952844619750977, 0.039495681762695314, 0.039472862243652346, 0.03946246337890625, 0.03936966323852539, 0.03958147048950195, 0.03954617691040039, 0.03978092956542969, 0.03971903991699219, 0.03956531143188476, 0.039831550598144534, 0.03990323257446289]",tokens/s,25.55306739425163,, @@ -12986,7 +12986,7 @@ ValueError: XGLMForCausalLM does not support an attention implementation through raise RuntimeError(f""Isolated process exited with non-zero code {isolated_process.exitcode}"") RuntimeError: Isolated process exited with non-zero code -9 ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v1-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,Deci/DeciCoder-1b,Deci/DeciCoder-1b,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v1-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,Deci/DeciCoder-1b,Deci/DeciCoder-1b,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,1,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -14416,7 +14416,7 @@ Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 4bit-gptq-exllama-v2-eager,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-1.3b,EleutherAI/pythia-1.3b,cuda,0,42,,,True,True,,float16,True,False,,eager,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.223-212.873.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.4.0,,4.44.2,,0.34.0,,,,1.21.4,,,,0.12.0,,,MB,1583.255552,1596.915712,0.0,1201.668096,1189.151232,s,1,8.466310546875,8.466310546875,0.0,8.466310546875,8.466310546875,8.466310546875,8.466310546875,[8.466310546875],,kWh,3.5574402879161464e-05,3.916985400328343e-06,1.096223099199961e-05,5.045361927148942e-05,,MB,1582.055424,1791.950848,0.0,1382.023168,1351.367168,s,10,0.47527958679199217,0.04752795867919922,0.0001405060291031286,0.0475042724609375,0.047698503112792966,0.04776477127075195,0.04781778579711914,"[0.04783103942871094, 0.047349502563476566, 0.0474101448059082, 0.04753100967407226, 0.047494945526123045, 0.04768377685546875, 0.04737731170654297, 0.04761840057373047, 0.047513599395751956, 0.04746985626220703]",tokens/s,5386.303285776069,kWh,1.4258651037194828e-06,1.5724679265124117e-07,9.432636272390402e-07,2.5263755236097645e-06,tokens/kWh,101330937.38741547,MB,1582.055424,1833.893888,0.0,1423.966208,1407.328256,s,10,13.680080322265626,1.3680080322265626,0.005092286023816497,1.3692230224609374,1.372664990234375,1.3746703247070313,1.3762745922851563,"[1.3663924560546874, 1.3722193603515624, 1.36963427734375, 1.3708880615234376, 1.3766756591796876, 1.3609461669921874, 1.3623831787109375, 1.368811767578125, 1.360443603515625, 1.371685791015625]",tokens/s,46.05236118201845,kWh,3.959619600836351e-05,4.367061092489081e-06,1.8069751583160582e-05,6.203300868401318e-05,tokens/kWh,1015588.3349284658,,s,630,13.677172939300538,0.02170979831635006,0.0003257925203109579,0.021650959968566894,0.02201076183319092,0.022220206356048582,0.023109855136871343,"[0.02183737564086914, 0.021971391677856444, 0.021774335861206053, 0.02169036865234375, 0.021478784561157226, 0.021575775146484375, 0.021658143997192382, 0.021591840744018556, 0.02197711944580078, 0.021448287963867187, 0.021748319625854492, 0.02164531135559082, 0.021510143280029297, 0.02131155204772949, 0.021342144012451172, 0.021348352432250976, 0.021510143280029297, 0.0214466552734375, 0.021825536727905274, 0.02328985595703125, 0.023052288055419923, 0.021985279083251954, 0.021737472534179687, 0.02159395217895508, 0.02135820770263672, 0.021760543823242186, 0.022798336029052735, 0.022199455261230468, 0.022010656356811525, 0.02200720024108887, 0.021852672576904295, 0.021866655349731444, 0.021757759094238282, 0.021776575088500977, 0.021420032501220702, 0.02141548728942871, 0.02150239944458008, 0.021568864822387696, 0.02147190475463867, 0.021398944854736326, 0.02151059150695801, 0.02261724853515625, 0.02175187110900879, 0.021832639694213868, 0.021522144317626953, 0.021604639053344726, 0.021640447616577147, 0.021680192947387697, 0.02156819152832031, 0.021274368286132814, 0.02143401527404785, 0.021611103057861326, 0.021487071990966798, 0.0213319034576416, 0.021312095642089843, 0.02142790412902832, 0.02147929573059082, 0.021328128814697266, 0.02131577682495117, 0.021321727752685548, 0.02128451156616211, 0.02140176010131836, 0.021917888641357422, 0.021655744552612304, 0.021626495361328126, 0.021563840866088868, 0.021434560775756836, 0.021370880126953123, 0.021405696868896484, 0.02141983985900879, 0.02174790382385254, 0.0214355525970459, 0.021416799545288086, 0.021501951217651367, 0.021476959228515623, 0.021436832427978517, 0.02153267288208008, 0.021452159881591798, 0.021387168884277344, 0.021381824493408204, 0.02141391944885254, 0.02147123146057129, 0.021594112396240234, 0.021640224456787108, 0.02175484848022461, 0.022253440856933593, 0.021506175994873047, 0.021579776763916016, 0.021671648025512694, 0.021647552490234374, 0.021651264190673827, 0.021673919677734375, 0.0216661434173584, 0.02167398452758789, 0.021784576416015625, 0.022504543304443358, 0.021859296798706053, 0.021761247634887695, 0.021699296951293946, 0.02184419250488281, 0.021999391555786132, 0.022024032592773437, 0.021909151077270508, 0.021924352645874022, 0.021831520080566408, 0.021915775299072265, 0.022086719512939453, 0.021868928909301758, 0.02190300750732422, 0.022864831924438476, 0.022024032592773437, 0.02205302429199219, 0.022237184524536133, 0.023139839172363282, 0.022139392852783202, 0.021809152603149414, 0.022282047271728514, 0.021975231170654298, 0.021981184005737304, 0.021725183486938478, 0.02167171287536621, 0.021833951950073243, 0.021721088409423828, 0.02184601593017578, 0.021690208435058592, 0.021622112274169922, 0.02175987243652344, 0.02165235137939453, 0.0216944637298584, 0.02162483215332031, 0.02165065574645996, 0.021822240829467772, 0.021741567611694337, 0.021651456832885742, 0.023347200393676756, 0.021571584701538086, 0.021628511428833007, 0.021868959426879882, 0.021728511810302734, 0.021762815475463868, 0.02168422317504883, 0.021634176254272462, 0.021855104446411134, 0.022321151733398437, 0.021559295654296876, 0.02191257667541504, 0.021799936294555664, 0.02167398452758789, 0.02168217658996582, 0.021618112564086914, 0.02167046356201172, 0.021621856689453125, 0.021588895797729494, 0.021708799362182618, 0.021550304412841798, 0.021771039962768555, 0.021788095474243162, 0.021668415069580078, 0.02167807960510254, 0.021939487457275392, 0.02173411178588867, 0.021538816452026367, 0.021575935363769533, 0.021631776809692384, 0.021573663711547852, 0.021658208847045897, 0.021582176208496093, 0.021749759674072267, 0.02162483215332031, 0.021790304183959962, 0.021707168579101564, 0.021611743927001954, 0.02174236869812012, 0.021544960021972655, 0.021747711181640626, 0.022042303085327147, 0.021905279159545897, 0.02189561653137207, 0.021768192291259765, 0.021710399627685548, 0.021640928268432617, 0.02163580894470215, 0.021563392639160156, 0.02168966484069824, 0.02205504035949707, 0.021524927139282227, 0.02162905693054199, 0.021596160888671875, 0.02168329620361328, 0.0215947208404541, 0.02175708770751953, 0.022016864776611328, 0.02207257652282715, 0.02200150489807129, 0.021978015899658202, 0.02183782386779785, 0.02178438377380371, 0.0217007999420166, 0.021542367935180665, 0.021638879776000975, 0.02160054397583008, 0.02154550361633301, 0.021573631286621094, 0.02167807960510254, 0.0220214729309082, 0.021695135116577148, 0.023735424041748047, 0.022834047317504883, 0.02168217658996582, 0.02164531135559082, 0.021594207763671876, 0.02172435188293457, 0.02173411178588867, 0.021724735260009766, 0.021698144912719725, 0.02151100730895996, 0.02159811210632324, 0.02150409507751465, 0.02153593635559082, 0.021566080093383788, 0.021452735900878907, 0.021502208709716798, 0.021549055099487305, 0.02160963249206543, 0.02166012763977051, 0.02234998321533203, 0.021653728485107424, 0.021590015411376954, 0.02162073516845703, 0.021624704360961915, 0.021708927154541015, 0.021655040740966795, 0.021715456008911133, 0.02171494483947754, 0.021648479461669923, 0.021617151260375975, 0.021598623275756835, 0.0216759033203125, 0.021869983673095703, 0.021912288665771485, 0.022030336380004883, 0.021542144775390626, 0.021825920104980467, 0.0216231689453125, 0.021612543106079102, 0.02159324836730957, 0.021603168487548827, 0.02166713523864746, 0.02168288040161133, 0.021784191131591798, 0.02188502311706543, 0.021725471496582032, 0.021899328231811524, 0.021958976745605468, 0.02193315124511719, 0.021725439071655275, 0.02178895950317383, 0.021696575164794923, 0.021952512741088868, 0.02390630340576172, 0.02294528007507324, 0.021631296157836915, 0.021466880798339843, 0.0219880313873291, 0.023131904602050782, 0.022017248153686525, 0.021768991470336913, 0.021746719360351562, 0.021492704391479493, 0.021491519927978514, 0.021562944412231444, 0.02178835105895996, 0.021662656784057616, 0.021710336685180662, 0.021647872924804686, 0.021536767959594725, 0.021513599395751953, 0.02174835205078125, 0.021776384353637695, 0.021596160888671875, 0.021501951217651367, 0.021493759155273438, 0.021559072494506837, 0.02167830467224121, 0.023006271362304688, 0.0216114559173584, 0.021579776763916016, 0.02164531135559082, 0.021702079772949218, 0.02160915184020996, 0.021518207550048827, 0.021615711212158203, 0.021676959991455077, 0.022882303237915038, 0.02233100891113281, 0.023055871963500976, 0.02203926467895508, 0.021883039474487304, 0.021764095306396485, 0.021585472106933595, 0.021685888290405273, 0.02164579200744629, 0.02167788887023926, 0.021867040634155274, 0.02187468719482422, 0.02191564750671387, 0.021695903778076172, 0.02154147148132324, 0.021574783325195312, 0.021640064239501954, 0.02166374397277832, 0.02166713523864746, 0.021545663833618164, 0.02189107131958008, 0.021618688583374023, 0.02166476821899414, 0.021589599609375, 0.021747968673706056, 0.021606559753417968, 0.021809152603149414, 0.02157948875427246, 0.021686368942260743, 0.021678272247314452, 0.02155628776550293, 0.021498815536499023, 0.021338111877441408, 0.02132905578613281, 0.021635936737060546, 0.02145894432067871, 0.021622783660888673, 0.021510143280029297, 0.021548383712768553, 0.021942943572998048, 0.021370880126953123, 0.021404991149902342, 0.021426111221313476, 0.021405887603759766, 0.02141241645812988, 0.021458303451538086, 0.021371519088745117, 0.02144576072692871, 0.021429119110107423, 0.02146713638305664, 0.02132294464111328, 0.021500736236572265, 0.021407743453979493, 0.021335136413574218, 0.021429119110107423, 0.0213668155670166, 0.021663455963134765, 0.021549215316772462, 0.021553279876708985, 0.021579776763916016, 0.021440256118774415, 0.021451007843017577, 0.021552255630493164, 0.021453695297241213, 0.021604352951049805, 0.02163711929321289, 0.021700607299804688, 0.021626880645751953, 0.021825536727905274, 0.022101503372192383, 0.021578239440917968, 0.021725183486938478, 0.02179465675354004, 0.02183945655822754, 0.02163974380493164, 0.02174950408935547, 0.021835647583007812, 0.02169385528564453, 0.021611488342285157, 0.021695903778076172, 0.021874847412109374, 0.022085567474365235, 0.021901023864746093, 0.02167478370666504, 0.021771488189697267, 0.02161664009094238, 0.021573631286621094, 0.02161430358886719, 0.02176790428161621, 0.021520959854125978, 0.02138105583190918, 0.021440576553344727, 0.021454368591308594, 0.02150217628479004, 0.02161484718322754, 0.021370655059814454, 0.021493247985839844, 0.021654239654541017, 0.021753856658935547, 0.02176630401611328, 0.02172297668457031, 0.02175584030151367, 0.02172115135192871, 0.021647327423095702, 0.021540735244750978, 0.021481632232666015, 0.021372928619384765, 0.02138057518005371, 0.02145948791503906, 0.021432319641113282, 0.021549055099487305, 0.021442047119140627, 0.021565792083740234, 0.021588031768798827, 0.021586015701293947, 0.021493759155273438, 0.02145020866394043, 0.021469024658203124, 0.02144326400756836, 0.02151219177246094, 0.021563392639160156, 0.02175116729736328, 0.021555360794067384, 0.021531103134155273, 0.021475040435791015, 0.021522432327270507, 0.021545248031616213, 0.021413888931274414, 0.021366432189941408, 0.021383520126342773, 0.02181065559387207, 0.02151683235168457, 0.021609760284423827, 0.021719263076782225, 0.021817440032958983, 0.022796703338623048, 0.021710687637329102, 0.02185641670227051, 0.021753856658935547, 0.02185420799255371, 0.021934080123901366, 0.021602304458618164, 0.02185625648498535, 0.021464256286621092, 0.02264147186279297, 0.02157513618469238, 0.021856128692626955, 0.02152284812927246, 0.021712928771972655, 0.021458623886108398, 0.0218338565826416, 0.021415456771850586, 0.02130179214477539, 0.021204416275024413, 0.021631872177124024, 0.0216944637298584, 0.02166579246520996, 0.022117759704589842, 0.02160089683532715, 0.021530624389648437, 0.021489664077758788, 0.021395456314086913, 0.02144256019592285, 0.021325824737548828, 0.021542911529541017, 0.021366336822509765, 0.021359039306640626, 0.02146303939819336, 0.021362688064575194, 0.021569536209106444, 0.0216494083404541, 0.021800960540771484, 0.02346931266784668, 0.02169523239135742, 0.022111360549926757, 0.0217445125579834, 0.02170457649230957, 0.021819520950317382, 0.022346912384033205, 0.022270816802978516, 0.021993471145629884, 0.022122047424316407, 0.02201171112060547, 0.023003776550292968, 0.02169856071472168, 0.02210383987426758, 0.021919776916503906, 0.021866016387939453, 0.021930784225463868, 0.021731199264526366, 0.021714080810546876, 0.02170966339111328, 0.02170863914489746, 0.02165692710876465, 0.021672767639160158, 0.021542207717895508, 0.02233318328857422, 0.021593023300170898, 0.021681312561035157, 0.02161248016357422, 0.021869024276733397, 0.021745439529418945, 0.021553823471069336, 0.021511808395385742, 0.021597600936889647, 0.02156755256652832, 0.02146601676940918, 0.02146713638305664, 0.021341728210449218, 0.021370687484741212, 0.021381792068481446, 0.021871423721313475, 0.021936128616333008, 0.021618495941162108, 0.02164508819580078, 0.021371295928955078, 0.021399072647094727, 0.021905887603759767, 0.021495136260986328, 0.021379743576049805, 0.021354496002197267, 0.02143756866455078, 0.021713792800903322, 0.02154489517211914, 0.021526592254638672, 0.021362688064575194, 0.021311487197875977, 0.021359712600708007, 0.02137321662902832, 0.021631616592407227, 0.022008064270019532, 0.021827295303344728, 0.02156470489501953, 0.021451520919799804, 0.02136796760559082, 0.021461759567260742, 0.021653568267822266, 0.021664928436279297, 0.021415807723999022, 0.021279743194580078, 0.021477632522583008, 0.02129280090332031, 0.021335552215576172, 0.02133247947692871, 0.02155913543701172, 0.021481632232666015, 0.021598207473754884, 0.021447999954223633, 0.021626752853393556, 0.021709535598754885, 0.0215184326171875, 0.02141168022155762, 0.021379104614257814, 0.02139334487915039, 0.021368511199951173, 0.02152057647705078, 0.02147056007385254, 0.021396448135375976, 0.021398624420166015, 0.021403968811035155, 0.021557151794433595, 0.02169107246398926, 0.021626688003540038, 0.021745376586914063, 0.021502431869506837, 0.02168009567260742, 0.022105663299560548, 0.022116832733154297, 0.022240959167480468, 0.02241708755493164, 0.022077375411987305, 0.021758176803588866, 0.021804800033569337, 0.0217890567779541, 0.02182102394104004, 0.021904191970825194, 0.022130495071411134, 0.022003007888793946, 0.021805376052856446, 0.021662080764770508, 0.021897216796875, 0.02156723213195801, 0.02151759910583496, 0.021486560821533204, 0.02182963180541992, 0.021821279525756836, 0.02198953628540039, 0.021612224578857423, 0.02170022392272949, 0.0216296329498291, 0.021696512222290038, 0.02159779167175293, 0.021603776931762696, 0.021598207473754884, 0.021826528549194337, 0.0220214729309082, 0.02184239959716797, 0.021799104690551758, 0.02206105613708496, 0.021798912048339843, 0.021661663055419923, 0.02184182357788086, 0.021885055541992188, 0.02166374397277832, 0.021987327575683592, 0.02187059211730957, 0.02214646339416504, 0.021583871841430666, 0.021760128021240235, 0.02177222442626953, 0.0216396484375, 0.021792255401611327, 0.021723712921142578, 0.022554239273071288, 0.022104352951049806, 0.021970016479492187, 0.021517087936401367, 0.021742944717407228, 0.02165657615661621, 0.02158729553222656, 0.021438880920410155, 0.0214836483001709, 0.021510112762451173, 0.021610143661499024, 0.021563776016235353, 0.02161664009094238, 0.02169968032836914, 0.02182032012939453, 0.021587968826293946, 0.021479040145874023, 0.021940223693847655, 0.021735807418823243, 0.021635072708129883, 0.022015615463256834, 0.022018367767333985, 0.021801023483276366, 0.021777599334716798]",tokens/s,46.06215062103461,, 4bit-gptq-exllama-v2-eager,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-1.4b,EleutherAI/pythia-1.4b,cuda,0,42,,,True,True,,float16,True,False,,eager,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.223-212.873.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.4.0,,4.44.2,,0.34.0,,,,1.21.4,,,,0.12.0,,,MB,1582.67392,1596.915712,0.0,1201.668096,1189.151232,s,1,8.203900390625,8.203900390625,0.0,8.203900390625,8.203900390625,8.203900390625,8.203900390625,[8.203900390625],,kWh,3.496767919166738e-05,3.849963228949563e-06,1.0954453207999426e-05,4.9772095628616364e-05,,MB,1564.16,1791.950848,0.0,1382.023168,1351.367168,s,10,0.47518137359619145,0.04751813735961914,0.0002168038865286773,0.04744075202941894,0.047669537353515624,0.04788238372802734,0.04805266082763672,"[0.04809523010253906, 0.04736966323852539, 0.04743475341796875, 0.04732377624511719, 0.04743775939941406, 0.04756444931030274, 0.047313503265380856, 0.04762223815917969, 0.04744374465942383, 0.047576255798339843]",tokens/s,5387.416557652121,kWh,1.426084696565e-06,1.5722692729590755e-07,9.489682388487577e-07,2.5322798627096657e-06,tokens/kWh,101094671.15773974,MB,1564.16,1833.893888,0.0,1423.966208,1407.328256,s,10,13.738055297851563,1.3738055297851564,0.003616846912428085,1.3729713134765624,1.378206689453125,1.3787253540039062,1.3791402856445312,"[1.3766226806640625, 1.3780914306640626, 1.3720076904296874, 1.369878662109375, 1.3776171875, 1.369964111328125, 1.371584228515625, 1.3739349365234375, 1.3792440185546875, 1.3691103515625]",tokens/s,45.85801893653194,kWh,3.954381329343452e-05,4.3613261565511e-06,1.8056352520951623e-05,6.196149197093724e-05,tokens/kWh,1016760.53942584,,s,630,13.735277492523181,0.02180202776590983,0.00033505996915220957,0.02172318458557129,0.022011130142211913,0.02230405149459839,0.023348841247558597,"[0.021792863845825194, 0.02171536064147949, 0.021740928649902343, 0.022104480743408202, 0.021720895767211913, 0.021602815628051757, 0.021725183486938478, 0.021708799362182618, 0.02162073516845703, 0.02167193603515625, 0.022177791595458983, 0.021825536727905274, 0.021644319534301758, 0.021639520645141602, 0.021813888549804688, 0.021712896347045898, 0.021765119552612306, 0.021570560455322265, 0.021763423919677734, 0.021709440231323242, 0.021766176223754884, 0.021712896347045898, 0.021567487716674806, 0.02166988754272461, 0.021839679718017577, 0.021624736785888672, 0.021889312744140625, 0.02546073532104492, 0.021745664596557617, 0.02179427146911621, 0.021656095504760744, 0.02175292778015137, 0.0224736328125, 0.02186182403564453, 0.02158799934387207, 0.021578271865844725, 0.021590015411376954, 0.022145023345947267, 0.023379743576049803, 0.022912351608276368, 0.02170524787902832, 0.021743967056274415, 0.02168217658996582, 0.021753631591796874, 0.021598432540893556, 0.02158736038208008, 0.02166204833984375, 0.021698816299438477, 0.021593376159667967, 0.021689056396484375, 0.021585376739501953, 0.021885120391845703, 0.02172323226928711, 0.021776639938354492, 0.021716991424560548, 0.021757280349731446, 0.021977247238159178, 0.02166124725341797, 0.021662656784057616, 0.021712608337402343, 0.021700895309448243, 0.02183782386779785, 0.021635072708129883, 0.02174835205078125, 0.021863935470581054, 0.021833728790283204, 0.021700895309448243, 0.02176041603088379, 0.021870399475097658, 0.02232121658325195, 0.021790847778320313, 0.021800960540771484, 0.021995519638061522, 0.022013952255249023, 0.021804927825927734, 0.021670015335083007, 0.02170639991760254, 0.021778783798217773, 0.021894943237304686, 0.021772512435913084, 0.021722623825073242, 0.021675552368164062, 0.021636064529418946, 0.021950304031372072, 0.02182467269897461, 0.02172313690185547, 0.021699583053588867, 0.02174937629699707, 0.021737855911254884, 0.02172835159301758, 0.021562271118164063, 0.02302297592163086, 0.022726463317871093, 0.021971744537353517, 0.021757984161376955, 0.021964448928833008, 0.021762399673461913, 0.021704704284667968, 0.021678016662597655, 0.02178873634338379, 0.02189107131958008, 0.02178278350830078, 0.021847007751464843, 0.02165020751953125, 0.02180860710144043, 0.02170275115966797, 0.02177043151855469, 0.021647296905517577, 0.02167977523803711, 0.021861024856567383, 0.021740959167480468, 0.021682559967041017, 0.021635295867919922, 0.021594112396240234, 0.02301692771911621, 0.021633567810058593, 0.021609888076782227, 0.021658208847045897, 0.02163043212890625, 0.0215614070892334, 0.02174745559692383, 0.023663328170776366, 0.02226585578918457, 0.02230067253112793, 0.02196463966369629, 0.021778079986572267, 0.02161177635192871, 0.021776351928710937, 0.02163167953491211, 0.021575935363769533, 0.021558687210083007, 0.02153660774230957, 0.021620960235595704, 0.021592096328735353, 0.021465599060058595, 0.021591615676879884, 0.021607999801635743, 0.023083295822143555, 0.022989408493041992, 0.02352332878112793, 0.02171494483947754, 0.021651168823242188, 0.021597824096679687, 0.02163983917236328, 0.02170675277709961, 0.021774335861206053, 0.021847583770751952, 0.021686752319335936, 0.021675039291381835, 0.021607391357421873, 0.021624544143676757, 0.021647552490234374, 0.021594112396240234, 0.02155939292907715, 0.021712896347045898, 0.02162073516845703, 0.021739295959472656, 0.02167625617980957, 0.02175939178466797, 0.021699167251586913, 0.022181407928466797, 0.02164169692993164, 0.021683616638183592, 0.021682783126831053, 0.021985279083251954, 0.022149120330810547, 0.021783903121948243, 0.021810943603515626, 0.021600576400756837, 0.02165190315246582, 0.021644927978515624, 0.02157948875427246, 0.02183247947692871, 0.02165558433532715, 0.021796863555908205, 0.021833728790283204, 0.021719039916992186, 0.02164531135559082, 0.021686111450195313, 0.021727392196655274, 0.021741567611694337, 0.021634687423706056, 0.021668224334716796, 0.021896575927734373, 0.02173388862609863, 0.021714431762695312, 0.021962976455688475, 0.021739839553833008, 0.021694368362426757, 0.021821632385253906, 0.021968832015991212, 0.021942527770996093, 0.021925535202026367, 0.021745311737060548, 0.021611200332641602, 0.0217488956451416, 0.02175062370300293, 0.02183919906616211, 0.021625503540039063, 0.02229043197631836, 0.021660896301269533, 0.021826175689697264, 0.02182547187805176, 0.02184582328796387, 0.021658016204833985, 0.021659263610839842, 0.02179110336303711, 0.02169036865234375, 0.021958656311035156, 0.02200371170043945, 0.02187446403503418, 0.021815519332885742, 0.02172313690185547, 0.021821056365966797, 0.021703039169311523, 0.021751808166503905, 0.02170639991760254, 0.02177039909362793, 0.021827775955200194, 0.021848031997680664, 0.021776256561279298, 0.02165478324890137, 0.02163715171813965, 0.02164620780944824, 0.021628192901611328, 0.021668127059936523, 0.021608896255493164, 0.021617759704589845, 0.02165648078918457, 0.02166374397277832, 0.02167193603515625, 0.021753856658935547, 0.021729280471801758, 0.021736991882324218, 0.021582304000854494, 0.02168012809753418, 0.021654687881469726, 0.021617504119873048, 0.021821216583251955, 0.021717216491699217, 0.021565439224243164, 0.021592063903808592, 0.021595680236816406, 0.02164963150024414, 0.02159436798095703, 0.021602304458618164, 0.021669408798217774, 0.021567968368530272, 0.021766143798828123, 0.021823007583618163, 0.021803487777709962, 0.0217509765625, 0.021703359603881835, 0.02184796714782715, 0.021753599166870117, 0.021706880569458006, 0.021842144012451173, 0.02162892723083496, 0.02188694381713867, 0.021833759307861328, 0.021786624908447266, 0.021665567398071288, 0.02167625617980957, 0.02269152069091797, 0.021626304626464844, 0.021735872268676758, 0.021807552337646485, 0.021638944625854493, 0.021737695693969727, 0.02171494483947754, 0.021702655792236326, 0.02164246368408203, 0.021635776519775392, 0.021640960693359374, 0.021758304595947266, 0.021859392166137696, 0.021659936904907227, 0.021680351257324218, 0.021682624816894532, 0.021702655792236326, 0.0216246395111084, 0.021647552490234374, 0.021708351135253906, 0.021815423965454103, 0.021800607681274415, 0.021651199340820312, 0.021666048049926757, 0.021796607971191408, 0.021805631637573243, 0.021721439361572267, 0.021719039916992186, 0.023518239974975586, 0.022268896102905274, 0.022314687728881837, 0.02189548873901367, 0.02169209671020508, 0.021776704788208007, 0.021862464904785155, 0.02280195236206055, 0.022339712142944335, 0.02185158348083496, 0.02192470359802246, 0.023119871139526366, 0.021951583862304686, 0.021842464447021485, 0.02233526420593262, 0.021801567077636717, 0.021747007369995117, 0.021731264114379884, 0.021822208404541014, 0.02186444854736328, 0.021693952560424806, 0.021674495697021484, 0.02170675277709961, 0.021650848388671876, 0.0215285758972168, 0.021638624191284178, 0.021666336059570312, 0.02191155242919922, 0.022071296691894532, 0.021729280471801758, 0.02182143974304199, 0.021962751388549806, 0.02219558334350586, 0.02194905662536621, 0.021573631286621094, 0.022183263778686523, 0.022041248321533202, 0.021810623168945314, 0.021717567443847657, 0.02151628875732422, 0.02166783905029297, 0.021630975723266603, 0.021551103591918946, 0.021606399536132814, 0.021725120544433593, 0.021624895095825197, 0.022220800399780274, 0.021761407852172853, 0.021645055770874024, 0.021658496856689455, 0.022054912567138672, 0.021857343673706054, 0.02160736083984375, 0.02161664009094238, 0.02165555191040039, 0.02163020706176758, 0.02169113540649414, 0.02168832015991211, 0.021745664596557617, 0.021581760406494142, 0.021643327713012694, 0.021544960021972655, 0.02166374397277832, 0.021656671524047853, 0.021617183685302733, 0.021739007949829102, 0.02171993637084961, 0.02175763130187988, 0.022328800201416015, 0.0216595516204834, 0.021586048126220704, 0.0216297607421875, 0.02161840057373047, 0.02173529624938965, 0.021710687637329102, 0.02169913673400879, 0.021722688674926757, 0.021744064331054688, 0.021610591888427736, 0.02164931106567383, 0.021600160598754883, 0.021573728561401367, 0.021700607299804688, 0.021753631591796874, 0.021758176803588866, 0.021740736007690428, 0.021823392868041993, 0.021796960830688477, 0.021600255966186522, 0.021719039916992186, 0.02209916877746582, 0.021606943130493165, 0.02166364860534668, 0.02247923278808594, 0.021751455307006836, 0.022157215118408204, 0.021613983154296874, 0.021647552490234374, 0.02169523239135742, 0.021692480087280273, 0.021607456207275392, 0.021865440368652345, 0.021879840850830078, 0.021788703918457032, 0.02175276756286621, 0.021710847854614256, 0.021811199188232423, 0.021740543365478517, 0.021724159240722657, 0.021958656311035156, 0.022087135314941407, 0.021795360565185547, 0.02202134323120117, 0.021977184295654296, 0.022110464096069336, 0.021756000518798828, 0.02187913513183594, 0.02189107131958008, 0.021702655792236326, 0.021755903244018555, 0.021766143798828123, 0.021985279083251954, 0.021724863052368162, 0.02176646423339844, 0.021811199188232423, 0.021853504180908204, 0.0216494083404541, 0.021686975479125976, 0.02164531135559082, 0.021593376159667967, 0.021752159118652345, 0.021612800598144532, 0.021571264266967774, 0.021659616470336915, 0.02160688018798828, 0.021768192291259765, 0.0220446720123291, 0.021542015075683593, 0.021588863372802733, 0.021660703659057617, 0.02201081657409668, 0.021806976318359376, 0.0216615047454834, 0.02158153533935547, 0.02165376091003418, 0.02155353546142578, 0.021618688583374023, 0.02164646339416504, 0.02154319953918457, 0.0216494083404541, 0.02179782485961914, 0.021716991424560548, 0.02162073516845703, 0.021647104263305662, 0.021686399459838867, 0.021678207397460937, 0.021831680297851562, 0.021622783660888673, 0.021610015869140624, 0.02162719917297363, 0.021780256271362305, 0.021837312698364256, 0.02161097526550293, 0.021682592391967775, 0.021817312240600586, 0.021935583114624024, 0.021770784378051758, 0.021773984909057617, 0.02175424003601074, 0.021764095306396485, 0.02171494483947754, 0.02165555191040039, 0.021774335861206053, 0.021754976272583007, 0.02180803108215332, 0.02189926338195801, 0.021746944427490235, 0.021836544036865236, 0.021635072708129883, 0.021882272720336913, 0.022131296157836915, 0.021833728790283204, 0.021743616104125976, 0.021890111923217773, 0.022125503540039063, 0.022128639221191407, 0.02175935935974121, 0.021770656585693358, 0.021820831298828124, 0.021713727951049804, 0.02169759941101074, 0.021704927444458007, 0.021946239471435546, 0.021916511535644532, 0.021958656311035156, 0.02187264060974121, 0.021985279083251954, 0.02190505599975586, 0.021789024353027344, 0.02174284744262695, 0.02170102310180664, 0.021731679916381835, 0.02234163284301758, 0.021845695495605468, 0.022308511734008787, 0.021745920181274414, 0.02169487953186035, 0.021780288696289063, 0.02175200080871582, 0.021767391204833984, 0.021616928100585936, 0.021826047897338868, 0.021772287368774415, 0.022202655792236327, 0.021602432250976564, 0.021749824523925782, 0.02171638488769531, 0.02166192054748535, 0.021701055526733397, 0.021741472244262695, 0.021755935668945313, 0.021791135787963867, 0.02211020851135254, 0.022345727920532226, 0.021646495819091796, 0.021586784362792967, 0.02159152030944824, 0.021940671920776367, 0.021757984161376955, 0.02211027145385742, 0.021766143798828123, 0.021819391250610352, 0.021757280349731446, 0.021650079727172852, 0.021721088409423828, 0.021719039916992186, 0.021708799362182618, 0.021708799362182618, 0.02181715202331543, 0.02179212760925293, 0.021663616180419922, 0.021805023193359373, 0.021819711685180664, 0.02245903968811035, 0.023595008850097656, 0.021786624908447266, 0.021882463455200195, 0.02186854362487793, 0.021884639739990233, 0.02163692855834961, 0.021741632461547852, 0.022858560562133787, 0.02177539253234863, 0.0216396484375, 0.021641727447509765, 0.021751808166503905, 0.02168009567260742, 0.021717023849487305, 0.02165551948547363, 0.02182761573791504, 0.02180271911621094, 0.021690656661987304, 0.0216494083404541, 0.02167807960510254, 0.021794815063476563, 0.021669792175292968, 0.021661792755126953, 0.021707807540893555, 0.02168726348876953, 0.022155263900756835, 0.021597824096679687, 0.02163283157348633, 0.021537343978881837, 0.022401023864746093, 0.022345567703247072, 0.02483216094970703, 0.022258880615234376, 0.02166864013671875, 0.021501983642578125, 0.02168627166748047, 0.021604352951049805, 0.021530624389648437, 0.021620031356811523, 0.0216746883392334, 0.021554496765136717, 0.02156819152832031, 0.021587968826293946, 0.021622783660888673, 0.021716991424560548, 0.021579776763916016, 0.02230681610107422, 0.021538400650024415, 0.023273183822631837, 0.021777088165283204, 0.021712799072265625, 0.021631071090698242, 0.021604223251342772, 0.02164748764038086, 0.021630975723266603, 0.02205081558227539, 0.0215817928314209, 0.021788703918457032, 0.021826656341552734, 0.021828128814697267, 0.021688703536987305, 0.021618623733520508, 0.0215982723236084, 0.021581760406494142, 0.021663808822631837, 0.021598207473754884, 0.021562528610229493, 0.021647712707519532, 0.02169465637207031, 0.0218668155670166, 0.021577728271484374, 0.021566848754882812, 0.021719680786132813, 0.02164121627807617, 0.02150809669494629, 0.0218024959564209, 0.021950143814086914, 0.02159062385559082, 0.021915712356567384, 0.0218175048828125, 0.021724512100219726, 0.021647615432739256, 0.021558944702148437, 0.02155801582336426, 0.021628320693969725, 0.021719072341918947, 0.021784160614013674, 0.021880800247192383, 0.02186342430114746, 0.021780479431152345, 0.021811199188232423, 0.0216693115234375, 0.02178060722351074, 0.021729248046875, 0.021746143341064453]",tokens/s,45.86729320488362,, -4bit-gptq-exllama-v2-eager,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,,float16,True,False,,eager,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v2-eager,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,,float16,True,False,,eager,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -14461,7 +14461,7 @@ ChildProcessError: Traceback (most recent call last): self._buffers[key] = fn(buf) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1160, in convert return t.to( -torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 70.00 MiB. GPU 0 has a total capacity of 14.74 GiB of which 16.12 MiB is free. Process 20361 has 14.72 GiB memory in use. Of the allocated memory 14.44 GiB is allocated by PyTorch, and 187.67 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 70.00 MiB. GPU 0 has a total capacity of 14.74 GiB of which 16.12 MiB is free. Process 20211 has 14.72 GiB memory in use. Of the allocated memory 14.44 GiB is allocated by PyTorch, and 187.67 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 4bit-gptq-exllama-v2-eager,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-12b,EleutherAI/pythia-12b,cuda,0,42,,,True,True,,float16,True,False,,eager,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.223-212.873.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.4.0,,4.44.2,,0.34.0,,,,1.21.4,,,,0.12.0,,,MB,6765.768704,7762.542592,0.0,7367.294976,7351.94368,s,1,12.484943359375,12.484943359375,0.0,12.484943359375,12.484943359375,12.484943359375,12.484943359375,[12.484943359375],,kWh,0.00015955133010832774,1.7592247732297916e-05,5.0381151416003384e-05,0.00022752472925662904,,MB,1654.890496,8372.813824,0.0,7962.886144,7872.44544,s,10,3.217915252685547,0.3217915252685547,0.0013950329344355533,0.3218920440673828,0.32337008056640626,0.32344077758789064,0.3234973352050781,"[0.3185030212402344, 0.32086541748046876, 0.32094903564453126, 0.3215930480957031, 0.3233543701171875, 0.3225209045410156, 0.32187457275390624, 0.3219095153808594, 0.3228338928222656, 0.323511474609375]",tokens/s,795.5461219382715,kWh,9.364775526823147e-06,1.032679123397765e-06,6.21373760987501e-06,1.6611192260095923e-05,tokens/kWh,15411295.949838202,MB,1666.101248,8624.472064,0.0,8214.544384,8118.577152,s,10,27.102155029296874,2.7102155029296875,0.0027258359209844218,2.709692626953125,2.712966064453125,2.7143250732421875,2.7154122802734375,"[2.71568408203125, 2.710020263671875, 2.7065537109375, 2.708927978515625, 2.708125732421875, 2.706820556640625, 2.712405517578125, 2.7126640625, 2.709364990234375, 2.711588134765625]",tokens/s,23.245383967399746,kWh,7.911330669484301e-05,8.72638364765581e-06,5.2567253859325075e-05,0.0001404069442018239,tokens/kWh,448695.7561688863,,s,630,27.098995811462423,0.043014279065813336,0.00037987361758398897,0.04298665618896484,0.04340562782287598,0.04354385929107666,0.044067572555542,"[0.04339251327514648, 0.04290611267089844, 0.042633216857910154, 0.04266726303100586, 0.04283404922485352, 0.04267686462402344, 0.04281257629394531, 0.043326400756835935, 0.04269865417480469, 0.042773696899414064, 0.04270108795166016, 0.04273596954345703, 0.04285414505004883, 0.04289318466186524, 0.042850879669189455, 0.04271641540527344, 0.04280915069580078, 0.04295999908447266, 0.043157215118408206, 0.042928031921386715, 0.04373324966430664, 0.042939937591552735, 0.042885536193847655, 0.04328995132446289, 0.04309030532836914, 0.04312297439575195, 0.042874462127685545, 0.04292959976196289, 0.04283491134643555, 0.04286054229736328, 0.04292607879638672, 0.04326399993896484, 0.04312995147705078, 0.0434246711730957, 0.04319232177734375, 0.04315750503540039, 0.042979328155517575, 0.043210014343261716, 0.043340511322021484, 0.04307353591918945, 0.043063297271728515, 0.043237377166748046, 0.04330825424194336, 0.04332998275756836, 0.043265472412109374, 0.04323612976074219, 0.04312409591674805, 0.0433191032409668, 0.04336304092407227, 0.04312841415405273, 0.0432624626159668, 0.04347097778320313, 0.043364673614501956, 0.043372222900390625, 0.04331958389282226, 0.04315017700195312, 0.043266654968261715, 0.04326019287109375, 0.04331315231323242, 0.043361793518066405, 0.04371916961669922, 0.0433172492980957, 0.043243743896484374, 0.043122783660888675, 0.04278262329101563, 0.04254515075683594, 0.04233420944213867, 0.042259807586669924, 0.04243059158325195, 0.04224873733520508, 0.04244895935058594, 0.042476478576660155, 0.04262400054931641, 0.04294451141357422, 0.04263705444335938, 0.04263935852050781, 0.04265497589111328, 0.04269801712036133, 0.04263497543334961, 0.04261798477172852, 0.04272995376586914, 0.04263468933105469, 0.04276649475097656, 0.042861278533935544, 0.04277193450927735, 0.043126815795898436, 0.043006561279296876, 0.04286435317993164, 0.042698879241943356, 0.04280876922607422, 0.04266876983642578, 0.042718494415283206, 0.04295139312744141, 0.043122432708740235, 0.04289152145385742, 0.042754207611083984, 0.04283564758300781, 0.0429323844909668, 0.042915233612060545, 0.04295363235473633, 0.04291657638549805, 0.04294345474243164, 0.043009151458740236, 0.04309491348266602, 0.04309811019897461, 0.04298115158081055, 0.043237598419189456, 0.04324147033691406, 0.04363673782348633, 0.04364716720581055, 0.0433928337097168, 0.044109825134277345, 0.04691558456420898, 0.04290259170532226, 0.043037151336669924, 0.04327881622314453, 0.04325312042236328, 0.04311011123657227, 0.04304790496826172, 0.04336019134521484, 0.043503456115722657, 0.043474815368652345, 0.04331862258911133, 0.04343190383911133, 0.0434442253112793, 0.04323984146118164, 0.04339212799072266, 0.042651905059814456, 0.04240614318847656, 0.042181087493896485, 0.04228496170043945, 0.0423765754699707, 0.04293222427368164, 0.0434054069519043, 0.04273001480102539, 0.04251238250732422, 0.04238655853271484, 0.04245939254760742, 0.042549888610839845, 0.042694591522216795, 0.04264556884765625, 0.042690464019775394, 0.04290569686889648, 0.04269875335693359, 0.042673728942871095, 0.04281388854980469, 0.04283353424072266, 0.04291945648193359, 0.04421283340454102, 0.04299174499511719, 0.04295897674560547, 0.04284620666503906, 0.04293632125854492, 0.04281865692138672, 0.04263753509521485, 0.04266463851928711, 0.04277043151855469, 0.04276838302612305, 0.04294374465942383, 0.0428223991394043, 0.04297119903564453, 0.04305500793457031, 0.04289750289916992, 0.04289056015014649, 0.04294105529785156, 0.042917728424072266, 0.04286223983764648, 0.04314729690551758, 0.04312063980102539, 0.043178462982177736, 0.0433983039855957, 0.04360665512084961, 0.04335638427734375, 0.04313087844848633, 0.04310835266113281, 0.043022560119628905, 0.04296384048461914, 0.04300601577758789, 0.04301091384887695, 0.043050430297851563, 0.04334988784790039, 0.04344083023071289, 0.04338483047485352, 0.04323904037475586, 0.043399551391601565, 0.043153408050537106, 0.04312396621704102, 0.0434672966003418, 0.04347312164306641, 0.04348108673095703, 0.04288409423828125, 0.042620990753173826, 0.042643455505371096, 0.042474559783935543, 0.042493824005126954, 0.04222390365600586, 0.042982559204101566, 0.04247814559936523, 0.042264575958251956, 0.04261580657958984, 0.042482688903808595, 0.04269465637207031, 0.043224414825439456, 0.04280963134765625, 0.04275971221923828, 0.04259254455566406, 0.042891841888427734, 0.04274176025390625, 0.04267734527587891, 0.04297001647949219, 0.04314835357666016, 0.04338275146484375, 0.043109344482421874, 0.04299760055541992, 0.04289737701416016, 0.04270918273925781, 0.042690208435058594, 0.04307388687133789, 0.04259635162353516, 0.042633216857910154, 0.04281897735595703, 0.04297296142578125, 0.04292076873779297, 0.042799102783203126, 0.04293609619140625, 0.0428812141418457, 0.04285776138305664, 0.042869503021240235, 0.04304012680053711, 0.043036670684814454, 0.04393971252441406, 0.04331292724609375, 0.04334662246704102, 0.0436778564453125, 0.04368396759033203, 0.04348281478881836, 0.04328656005859375, 0.043423168182373045, 0.043213119506835936, 0.04329033660888672, 0.04339206314086914, 0.043089664459228516, 0.04297283172607422, 0.04301017761230469, 0.043087135314941405, 0.04310521697998047, 0.0430544319152832, 0.04329539108276367, 0.0432509765625, 0.043344097137451174, 0.04339548873901367, 0.043599967956542966, 0.043315361022949216, 0.04273196792602539, 0.04244416046142578, 0.0423985595703125, 0.04230579376220703, 0.04235033416748047, 0.04253900909423828, 0.04252492904663086, 0.04249484634399414, 0.04268064117431641, 0.04270153427124023, 0.04263673782348633, 0.0426530876159668, 0.04245743942260742, 0.04241052627563477, 0.04261471939086914, 0.04282294464111328, 0.04270336151123047, 0.042927936553955076, 0.043076416015625, 0.04304851150512695, 0.04296300888061524, 0.04290166473388672, 0.04305920028686523, 0.04289712142944336, 0.04295635223388672, 0.04283583831787109, 0.04289827346801758, 0.04281753540039063, 0.042799102783203126, 0.042774177551269534, 0.04262947082519531, 0.04281753540039063, 0.042995712280273435, 0.04292812728881836, 0.042981342315673826, 0.042939678192138675, 0.04293452835083008, 0.043326976776123044, 0.043121662139892575, 0.04309811019897461, 0.044436702728271486, 0.04318611145019531, 0.04359219360351563, 0.04317219161987305, 0.04318003082275391, 0.0431769905090332, 0.04339811325073242, 0.04340035247802734, 0.04325667190551758, 0.04330624008178711, 0.04327091217041015, 0.04319145584106445, 0.04305599975585937, 0.04314716720581055, 0.043165760040283205, 0.04337238311767578, 0.04333580780029297, 0.04356304168701172, 0.04319622421264648, 0.043141311645507815, 0.04329395294189453, 0.04348169708251953, 0.04351996612548828, 0.042734527587890626, 0.0424672966003418, 0.04235001754760742, 0.04251055908203125, 0.04264550399780274, 0.04276176071166992, 0.04258095932006836, 0.04257756805419922, 0.04234377670288086, 0.04252713775634766, 0.042531265258789065, 0.042461406707763674, 0.04265167999267578, 0.042692352294921875, 0.0427613754272461, 0.042576416015625, 0.04268012619018555, 0.042706497192382814, 0.04284425735473633, 0.042871646881103516, 0.042856449127197264, 0.04312844848632812, 0.04304115295410156, 0.04301728057861328, 0.04295961761474609, 0.042829856872558594, 0.04280073547363281, 0.04336838531494141, 0.042945152282714845, 0.042942272186279294, 0.042750175476074216, 0.042774494171142576, 0.04271104049682617, 0.04279276657104492, 0.04293036651611328, 0.042985183715820316, 0.04294646453857422, 0.042871166229248046, 0.043028480529785154, 0.04294451141357422, 0.04311859130859375, 0.043259552001953125, 0.043964126586914065, 0.043610305786132814, 0.04332793426513672, 0.04339199829101562, 0.04333465576171875, 0.0432470703125, 0.043098464965820316, 0.04309811019897461, 0.043378273010253904, 0.043190113067626955, 0.04317068862915039, 0.04312255859375, 0.04317593765258789, 0.04312390518188477, 0.04307254409790039, 0.043183902740478515, 0.04321673583984375, 0.043149471282958984, 0.04325900650024414, 0.04354547119140625, 0.04343603134155274, 0.043044864654541014, 0.04262911987304688, 0.042618881225585936, 0.04242432022094727, 0.04251155090332031, 0.04255007934570312, 0.04272700881958008, 0.042664352416992186, 0.042675392150878906, 0.042658622741699216, 0.04277679824829102, 0.042898944854736325, 0.04281689453125, 0.042703777313232424, 0.0433438720703125, 0.04309161758422852, 0.04275439834594726, 0.04297926330566406, 0.04299782562255859, 0.042858558654785155, 0.043007232666015624, 0.04306208038330078, 0.0430365104675293, 0.043097248077392576, 0.042957118988037106, 0.042900032043457034, 0.042821632385253904, 0.04264755249023437, 0.042681888580322264, 0.04283391952514649, 0.04318435287475586, 0.04304054260253906, 0.042872512817382816, 0.04275609588623047, 0.04273017501831055, 0.042807392120361325, 0.04308777618408203, 0.04311843109130859, 0.04308812713623047, 0.04323680114746094, 0.04336627197265625, 0.04362924957275391, 0.04319427108764649, 0.0433787841796875, 0.04427775955200195, 0.04339244842529297, 0.04331577682495117, 0.043205760955810545, 0.04380070495605469, 0.043447040557861326, 0.043227134704589845, 0.04311859130859375, 0.04305100631713867, 0.04318790435791016, 0.04325203323364258, 0.043003902435302735, 0.04312441635131836, 0.04326041412353516, 0.04318598556518555, 0.04331267166137695, 0.043326976776123044, 0.04392035293579102, 0.04344841766357422, 0.04281520080566406, 0.04252467346191406, 0.04247836685180664, 0.04235212707519531, 0.04245532989501953, 0.042557823181152345, 0.042552513122558595, 0.04264432144165039, 0.042547168731689455, 0.04255744171142578, 0.042593441009521484, 0.042514686584472654, 0.04277104187011719, 0.04268147277832031, 0.04259929656982422, 0.04274748611450195, 0.042839969635009766, 0.042888702392578124, 0.04286975860595703, 0.04295206451416016, 0.042945152282714845, 0.043060543060302735, 0.042961631774902344, 0.042874462127685545, 0.0429714241027832, 0.04287088012695312, 0.043012096405029294, 0.04292227172851563, 0.042979198455810545, 0.04278572845458985, 0.04279004669189453, 0.04276172637939453, 0.042753280639648436, 0.04296214294433594, 0.04308505630493164, 0.04327062225341797, 0.04319347381591797, 0.04313932800292969, 0.04369887924194336, 0.04321279907226563, 0.043235294342041014, 0.04354188919067383, 0.04347561645507812, 0.043407615661621095, 0.043398944854736325, 0.04346262359619141, 0.0434442253112793, 0.04338687896728516, 0.04336608123779297, 0.04365548706054687, 0.04322844696044922, 0.043254016876220706, 0.043208351135253904, 0.04333663940429688, 0.04348483276367188, 0.04350950241088867, 0.043557342529296876, 0.04349747085571289, 0.043401214599609376, 0.04358473587036133, 0.043574047088623044, 0.04361833572387695, 0.043200416564941405, 0.04267948913574219, 0.04248617553710937, 0.04257388687133789, 0.04260892868041992, 0.042546688079833986, 0.042656383514404296, 0.042503807067871095, 0.04257628631591797, 0.04242147064208984, 0.042463134765625, 0.04256835174560547, 0.04272355270385742, 0.04278217697143555, 0.04293072128295899, 0.042805248260498044, 0.0428807373046875, 0.043251232147216795, 0.042922592163085936, 0.04285660934448242, 0.04308992004394531, 0.043175838470458985, 0.043081825256347656, 0.0432735366821289, 0.04299436950683594, 0.042833759307861326, 0.04280131149291992, 0.04268841552734375, 0.042749248504638675, 0.04283689498901367, 0.04282969665527344, 0.04277967834472656, 0.04297808074951172, 0.04299766540527344, 0.042842174530029295, 0.04284438323974609, 0.04301801681518555, 0.04305667114257813, 0.04286124801635742, 0.04293632125854492, 0.04308515167236328, 0.043240097045898436, 0.04328646469116211, 0.04334393692016601, 0.04345187377929687, 0.04347123336791992, 0.043284160614013675, 0.043159839630126956, 0.04311648178100586, 0.044163326263427734, 0.04320857620239258, 0.04326518249511719, 0.04327657699584961, 0.043313343048095705, 0.04326383972167969, 0.043231903076171876, 0.043218238830566406, 0.04324739074707031, 0.043192607879638675, 0.043342464447021486, 0.0432988166809082, 0.043225086212158204, 0.04327324676513672, 0.04348108673095703, 0.042967041015625, 0.042674175262451174, 0.04249151992797852, 0.04243084716796875, 0.042240001678466796, 0.04242371368408203, 0.042477535247802733, 0.042566272735595705, 0.0425615348815918, 0.042909183502197266, 0.04448470306396484, 0.04250848007202149, 0.04247087860107422, 0.042474239349365235, 0.04253488159179687, 0.04274998474121094, 0.04283596801757812, 0.042894432067871094, 0.04296099090576172, 0.04316447830200195, 0.043119873046875, 0.043125503540039065, 0.04311859130859375, 0.0432125129699707, 0.04272911834716797, 0.042740352630615236, 0.042782718658447266, 0.04290505599975586, 0.04291619110107422, 0.04298553466796875, 0.04299993515014648, 0.04304256057739258, 0.042987777709960935, 0.042907615661621094, 0.04304899215698242, 0.042874881744384766, 0.04354048156738281, 0.043030143737792965, 0.043358592987060546, 0.043412830352783205, 0.04321267318725586, 0.04331302261352539, 0.043273120880126956, 0.043401214599609376, 0.04361612701416016, 0.04336579132080078, 0.04337148666381836, 0.04324121475219726, 0.04315702438354492, 0.042947040557861325, 0.04310806274414063, 0.043299423217773435, 0.04313631820678711, 0.04316128158569336, 0.04308819198608398, 0.04321446228027344, 0.043071231842041015, 0.0432281608581543, 0.04335520172119141, 0.043235679626464844, 0.043499393463134764, 0.04387311935424805]",tokens/s,23.24809392876178,, @@ -14477,7 +14477,7 @@ torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 70.00 MiB. GPU 0 h raise RuntimeError(f""Isolated process exited with non-zero code {isolated_process.exitcode}"") RuntimeError: Isolated process exited with non-zero code -9 ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v2-eager,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,Deci/DeciCoder-1b,Deci/DeciCoder-1b,cuda,0,42,,,True,True,,float16,True,False,,eager,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v2-eager,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,Deci/DeciCoder-1b,Deci/DeciCoder-1b,cuda,0,42,,,True,True,,float16,True,False,,eager,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -14865,7 +14865,7 @@ ValueError: OPTForCausalLM does not support an attention implementation through ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 4bit-gptq-exllama-v2-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,stabilityai/stablelm-base-alpha-7b,stabilityai/stablelm-base-alpha-7b,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,,MB,4913.119232,5995.626496,0.0,5593.104384,5582.857216,s,1,11.47325390625,11.47325390625,0.0,11.47325390625,11.47325390625,11.47325390625,11.47325390625,[11.47325390625],,kWh,0.00011755960349584407,1.2960246374984289e-05,3.553197286999821e-05,0.00016605182274082657,,MB,1562.48064,6324.87936,0.0,5907.677184,5844.559872,s,10,2.019981018066406,0.20199810180664063,0.0003855504032527179,0.2019721450805664,0.20252928771972656,0.2026807891845703,0.2028019903564453,"[0.20204623413085937, 0.2017305908203125, 0.2015630340576172, 0.2019661407470703, 0.2019781494140625, 0.20283229064941405, 0.20159686279296876, 0.20210099792480468, 0.20249562072753907, 0.20167109680175782]",tokens/s,1267.3386418504656,kWh,5.911116723167197e-06,6.518912872977685e-07,3.936292037920175e-06,1.049930004838514e-05,tokens/kWh,24382577.77377973,MB,1570.7136,6324.87936,0.0,5907.677184,5844.562432,s,10,14.89311328125,1.4893113281250003,0.0027831900490972463,1.4886818237304689,1.4936528686523438,1.494207843017578,1.4946518225097656,"[1.4863668212890624, 1.4868624267578125, 1.4947628173828125, 1.48802294921875, 1.4907061767578125, 1.4872960205078125, 1.4896527099609376, 1.4865731201171875, 1.4893406982421875, 1.493529541015625]",tokens/s,42.30143074202972,kWh,4.347846238225187e-05,4.795396007565931e-06,2.8716400750879446e-05,7.699025914069724e-05,tokens/kWh,818285.3350950477,,s,630,14.89045707130433,0.023635646144927497,0.0001889448159364638,0.023624704360961916,0.023861656951904297,0.023941408443450926,0.02410854320526123,"[0.02360019111633301, 0.023313343048095705, 0.023403615951538087, 0.023172128677368165, 0.023213951110839844, 0.02340220832824707, 0.023537376403808593, 0.02366249656677246, 0.023577247619628906, 0.023399839401245116, 0.023212640762329102, 0.02330419158935547, 0.023470207214355467, 0.023437183380126955, 0.023481760025024414, 0.023508703231811524, 0.023382656097412108, 0.02344576072692871, 0.023635967254638672, 0.023643999099731444, 0.023584928512573242, 0.023627872467041015, 0.023506847381591797, 0.02352332878112793, 0.02373788833618164, 0.023730655670166016, 0.02353971290588379, 0.023431167602539063, 0.02333695983886719, 0.02328166389465332, 0.023473407745361326, 0.023642719268798826, 0.02379520034790039, 0.023833248138427736, 0.023554048538208007, 0.023443456649780273, 0.023747840881347657, 0.024095487594604493, 0.023625600814819337, 0.02351852798461914, 0.023407424926757812, 0.023475839614868165, 0.023800031661987305, 0.02380611228942871, 0.02386147117614746, 0.023856576919555665, 0.023721599578857423, 0.023631935119628907, 0.024013248443603516, 0.023720287322998048, 0.023556192398071288, 0.02345919990539551, 0.023355424880981444, 0.023673215866088868, 0.023803520202636718, 0.023804288864135742, 0.023771135330200196, 0.02376006317138672, 0.02379244804382324, 0.023848960876464844, 0.023748287200927733, 0.023734592437744142, 0.02369740867614746, 0.023597984313964843, 0.02326655960083008, 0.023408863067626955, 0.02353001594543457, 0.023623712539672853, 0.02346937561035156, 0.023420991897583007, 0.023335519790649413, 0.023435264587402343, 0.023373023986816406, 0.023568607330322264, 0.023640640258789064, 0.023534751892089843, 0.023407455444335937, 0.023576576232910155, 0.02366873550415039, 0.023778751373291017, 0.023627744674682618, 0.023528032302856446, 0.023590911865234376, 0.023533344268798828, 0.023586912155151366, 0.023407936096191406, 0.023497535705566407, 0.023408639907836915, 0.023574047088623047, 0.023470752716064452, 0.023682880401611327, 0.023827455520629884, 0.023774208068847655, 0.02352332878112793, 0.02357811164855957, 0.023828512191772462, 0.023793983459472656, 0.023666175842285156, 0.023585439682006836, 0.023382272720336914, 0.023448352813720704, 0.023632991790771486, 0.023648128509521485, 0.02360927963256836, 0.023668863296508788, 0.02342086410522461, 0.02351103973388672, 0.0236810245513916, 0.023633920669555664, 0.023676895141601564, 0.0235947208404541, 0.023449920654296876, 0.02347113609313965, 0.02366281509399414, 0.02367692756652832, 0.023650751113891602, 0.02358940887451172, 0.023436992645263673, 0.02379132843017578, 0.02392019271850586, 0.023984960556030274, 0.02391766357421875, 0.023591840744018554, 0.02366054344177246, 0.023949312210083007, 0.023799072265625, 0.023705631256103515, 0.023591680526733397, 0.02364022445678711, 0.02374995231628418, 0.023755456924438475, 0.02380905532836914, 0.023835615158081056, 0.023729663848876953, 0.023509504318237305, 0.023721664428710938, 0.023781440734863282, 0.023767200469970703, 0.02370947265625, 0.02356870460510254, 0.023380096435546876, 0.0235231990814209, 0.02378246307373047, 0.02376799964904785, 0.0236810245513916, 0.023435264587402343, 0.023463712692260743, 0.023386335372924803, 0.023557952880859375, 0.02349484825134277, 0.02343926429748535, 0.02351091194152832, 0.023699840545654296, 0.02369094467163086, 0.0237938232421875, 0.023948448181152344, 0.024212255477905273, 0.023611455917358398, 0.02360220718383789, 0.0238067512512207, 0.023883968353271483, 0.02387513542175293, 0.023820735931396483, 0.023564287185668945, 0.0235948486328125, 0.023737920761108398, 0.023765087127685547, 0.023736543655395508, 0.02359939193725586, 0.023513088226318358, 0.02356425666809082, 0.023909439086914064, 0.023972831726074218, 0.02378153610229492, 0.02362303924560547, 0.02354979133605957, 0.023880319595336916, 0.024153375625610353, 0.023853599548339845, 0.02372969627380371, 0.023458463668823242, 0.0235020809173584, 0.023849727630615235, 0.024155296325683594, 0.024111967086791992, 0.023946943283081053, 0.023822528839111328, 0.02393267250061035, 0.023902496337890624, 0.02365190315246582, 0.02356268882751465, 0.023549184799194336, 0.02374924850463867, 0.023947391510009765, 0.023767040252685546, 0.023618623733520507, 0.023572864532470702, 0.0235850887298584, 0.023527679443359376, 0.023572608947753905, 0.02350067138671875, 0.023517183303833008, 0.023413951873779298, 0.02326406478881836, 0.02333443260192871, 0.02343075180053711, 0.02338899230957031, 0.023396352767944335, 0.023369184494018556, 0.023411296844482423, 0.023393695831298827, 0.023556608200073242, 0.02352547264099121, 0.023529472351074218, 0.023590911865234376, 0.02365235137939453, 0.02368060874938965, 0.023614879608154296, 0.024060640335083008, 0.023832128524780272, 0.02360188865661621, 0.023377599716186522, 0.023427200317382813, 0.02357267189025879, 0.023723583221435546, 0.023615840911865235, 0.023431264877319335, 0.023331071853637694, 0.02344633674621582, 0.023705951690673827, 0.02355583953857422, 0.023533632278442383, 0.02347667121887207, 0.023420608520507813, 0.0235567684173584, 0.023879680633544922, 0.023801855087280274, 0.023748832702636717, 0.02366841506958008, 0.02369340705871582, 0.023825855255126954, 0.024011327743530275, 0.02400444793701172, 0.023808160781860353, 0.023807487487792968, 0.023699968338012696, 0.02385305595397949, 0.024031007766723633, 0.023799776077270508, 0.02364031982421875, 0.0235797119140625, 0.023540576934814452, 0.023525503158569334, 0.023291231155395508, 0.023364383697509764, 0.023504287719726562, 0.023420480728149413, 0.023325471878051757, 0.023333919525146483, 0.02336832046508789, 0.023523168563842775, 0.0234005126953125, 0.023383968353271483, 0.02334979248046875, 0.023741695404052736, 0.023921215057373046, 0.02388528060913086, 0.023886112213134764, 0.02352707290649414, 0.023374624252319336, 0.023629823684692384, 0.023764991760253908, 0.023700544357299805, 0.02372822380065918, 0.023640735626220703, 0.02377676773071289, 0.023851167678833007, 0.023912031173706053, 0.02375315284729004, 0.023577247619628906, 0.02333171272277832, 0.023619903564453124, 0.023763776779174805, 0.02378940773010254, 0.023759071350097655, 0.02373200035095215, 0.023545087814331053, 0.023831296920776367, 0.023838272094726564, 0.02373881530761719, 0.023549663543701173, 0.023501216888427736, 0.02346544075012207, 0.023715551376342774, 0.02382099151611328, 0.024028480529785155, 0.023799903869628908, 0.023644767761230468, 0.023430624008178712, 0.023568992614746095, 0.023633440017700194, 0.023631296157836913, 0.023704320907592773, 0.023599456787109375, 0.023746143341064452, 0.023889471054077148, 0.02386921691894531, 0.02393600082397461, 0.023805152893066405, 0.02361622428894043, 0.023734111785888672, 0.023912607192993166, 0.02381772804260254, 0.023871999740600586, 0.02372559928894043, 0.02374838447570801, 0.023527040481567382, 0.023513824462890624, 0.02364009666442871, 0.02354774475097656, 0.023307872772216798, 0.023384479522705077, 0.023434848785400392, 0.0234903678894043, 0.023454303741455077, 0.023461856842041016, 0.02330403137207031, 0.02335670471191406, 0.023445951461791993, 0.023406463623046873, 0.023347808837890626, 0.023243871688842774, 0.023200576782226562, 0.023230560302734377, 0.023508640289306642, 0.023589216232299804, 0.02353561592102051, 0.023640064239501952, 0.023571744918823242, 0.023380704879760742, 0.023607007980346678, 0.023636255264282226, 0.02370969581604004, 0.02357072067260742, 0.02338323211669922, 0.02320368003845215, 0.023611679077148437, 0.023730592727661134, 0.023853023529052733, 0.02359823989868164, 0.023542591094970703, 0.023355552673339844, 0.023539199829101562, 0.02371625518798828, 0.02376870346069336, 0.02385081672668457, 0.02376252746582031, 0.02354470443725586, 0.023775232315063476, 0.02399216079711914, 0.02394463920593262, 0.023928991317749025, 0.0237861442565918, 0.023590080261230467, 0.023822944641113283, 0.023939296722412108, 0.023917760848999024, 0.023839487075805663, 0.023529407501220703, 0.023740447998046876, 0.023787328720092774, 0.02378371238708496, 0.02388787269592285, 0.02368716812133789, 0.023576480865478516, 0.0236661434173584, 0.02379840087890625, 0.023796735763549806, 0.023577024459838867, 0.023066591262817383, 0.023328895568847655, 0.023549951553344727, 0.023762943267822266, 0.02370150375366211, 0.02357219123840332, 0.023464256286621094, 0.023385919570922852, 0.023656415939331054, 0.023639488220214843, 0.023627872467041015, 0.023617855072021486, 0.02345385551452637, 0.023548095703125, 0.023789567947387694, 0.023760896682739258, 0.023569536209106446, 0.023593568801879884, 0.023451936721801757, 0.023369279861450196, 0.02351763153076172, 0.023621631622314454, 0.02360633659362793, 0.023482431411743165, 0.023310720443725588, 0.023351808547973633, 0.023584320068359376, 0.023599327087402342, 0.023753023147583006, 0.02382022476196289, 0.02394313621520996, 0.023571456909179687, 0.02373843193054199, 0.02377209663391113, 0.02368921661376953, 0.023476224899291992, 0.023343103408813477, 0.023508256912231445, 0.023784160614013672, 0.02371788787841797, 0.023762144088745118, 0.023776031494140624, 0.023614719390869142, 0.023717920303344728, 0.023775039672851564, 0.023703775405883788, 0.0237063045501709, 0.023518976211547853, 0.023388416290283202, 0.023606784820556642, 0.02381875228881836, 0.02383452796936035, 0.023760896682739258, 0.02370159912109375, 0.02372198486328125, 0.024020992279052734, 0.02408176040649414, 0.024019615173339844, 0.023797183990478515, 0.023599679946899415, 0.02382636833190918, 0.0239484806060791, 0.023662879943847658, 0.023375551223754884, 0.023371807098388674, 0.023513088226318358, 0.023775232315063476, 0.023744287490844725, 0.023511264801025392, 0.023547391891479492, 0.023323135375976564, 0.023416831970214845, 0.023509151458740236, 0.02351638412475586, 0.023602943420410156, 0.02374950408935547, 0.023453535079956053, 0.023390335083007814, 0.023478303909301758, 0.023439136505126953, 0.02345974349975586, 0.023449663162231446, 0.023574655532836913, 0.02325721549987793, 0.023201791763305665, 0.02349056053161621, 0.02349180793762207, 0.023505311965942383, 0.02350102424621582, 0.023385696411132813, 0.023435808181762694, 0.023775264739990233, 0.023994367599487306, 0.02372140884399414, 0.023558624267578127, 0.023470399856567382, 0.0235314884185791, 0.023529279708862306, 0.02354764747619629, 0.02367862319946289, 0.023591520309448243, 0.023476224899291992, 0.02356150436401367, 0.02359107208251953, 0.02367964744567871, 0.023704479217529297, 0.02350796890258789, 0.023601055145263672, 0.023603263854980468, 0.02365222358703613, 0.023695520401000977, 0.0236810245513916, 0.02361452865600586, 0.023534400939941406, 0.023623807907104492, 0.02386534309387207, 0.02383407974243164, 0.023845504760742188, 0.02414806365966797, 0.023631584167480468, 0.0238056640625, 0.023840608596801757, 0.023787935256958007, 0.023789695739746094, 0.0236844482421875, 0.023730367660522462, 0.023545663833618165, 0.02346598434448242, 0.02350694465637207, 0.023573984146118165, 0.023338720321655272, 0.023563072204589842, 0.023924736022949217, 0.0237871036529541, 0.023583232879638674, 0.023483360290527344, 0.023374624252319336, 0.02338364791870117, 0.023484928131103516, 0.023486528396606445, 0.023516511917114256, 0.02343199920654297, 0.023322656631469728, 0.02340640068054199, 0.023572576522827147, 0.023562143325805664, 0.023639680862426758, 0.023564672470092772, 0.023646207809448243, 0.023932191848754884, 0.023656543731689454, 0.023734592437744142, 0.023674688339233398, 0.023526079177856447, 0.023585599899291994, 0.023813087463378905, 0.02386332893371582, 0.023926015853881835, 0.023733152389526366, 0.02351702308654785, 0.02348441505432129, 0.023688352584838868, 0.023669439315795897, 0.023621055603027345, 0.02359369659423828, 0.02345939254760742, 0.02339232063293457, 0.023531551361083983, 0.023525728225708007, 0.02357151985168457, 0.023704544067382812, 0.02355836868286133, 0.023666336059570314, 0.02387984085083008, 0.02384889602661133, 0.023785472869873047, 0.023713024139404296, 0.02358348846435547, 0.023854719161987305, 0.02380633544921875, 0.02385686492919922, 0.023742752075195314, 0.023554048538208007, 0.023467296600341796, 0.023689952850341797, 0.02403740882873535, 0.024037343978881836, 0.023887264251708985, 0.023525087356567383, 0.02338240051269531, 0.023764991760253908, 0.023823455810546876, 0.023689695358276367, 0.02373049545288086, 0.023594207763671875, 0.02363702392578125, 0.023883295059204102, 0.023814495086669923, 0.023731903076171876, 0.023578943252563475, 0.023460063934326172, 0.02360655975341797, 0.023691007614135742, 0.023765472412109374, 0.023660831451416016, 0.02367910385131836, 0.023567520141601562, 0.023648223876953124, 0.02375142478942871, 0.023619583129882812, 0.023611391067504883, 0.023410911560058593, 0.02340003204345703, 0.023681215286254883, 0.023813888549804686, 0.023799871444702147, 0.023840831756591796, 0.02358518409729004, 0.023695072174072265, 0.02383228874206543, 0.0238123836517334, 0.023666688919067383, 0.02364313507080078, 0.02338515281677246, 0.023449535369873046, 0.023814144134521483, 0.023783231735229494, 0.023711040496826173, 0.023701536178588868, 0.02355830383300781, 0.02366752052307129, 0.02410016059875488, 0.02418659210205078, 0.024014911651611327, 0.023770111083984375, 0.0236824951171875, 0.023593311309814454, 0.023789567947387694, 0.023837696075439452, 0.023694015502929686, 0.02374185562133789, 0.023705984115600588, 0.023877311706542968, 0.02456889533996582, 0.023727359771728514, 0.023574176788330077, 0.02340656089782715, 0.023460031509399414, 0.023722719192504883, 0.02373420715332031, 0.023666751861572265]",tokens/s,42.30897661389352,, -4bit-gptq-exllama-v2-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neo,EleutherAI/gpt-neo-2.7B,EleutherAI/gpt-neo-2.7B,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v2-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neo,EleutherAI/gpt-neo-2.7B,EleutherAI/gpt-neo-2.7B,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -15003,7 +15003,7 @@ ChildProcessError: Traceback (most recent call last): ValueError: The model class you are passing has a `config_class` attribute that is not consistent with the config class you passed (model has and you passed . Fix one of those so they match! ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v2-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,deci,Deci/DeciLM-7B,Deci/DeciLM-7B,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v2-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,deci,Deci/DeciLM-7B,Deci/DeciLM-7B,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -15111,7 +15111,7 @@ ChildProcessError: Traceback (most recent call last): ValueError: RecurrentGemmaForCausalLM does not support an attention implementation through torch.nn.functional.scaled_dot_product_attention yet. Please request the support for this architecture: https://github.com/huggingface/transformers/issues/28005. If you believe this error is a bug, please open an issue in Transformers GitHub repository and load your model with the argument `attn_implementation=""eager""` meanwhile. Example: `model = AutoModel.from_pretrained(""openai/whisper-tiny"", attn_implementation=""eager"")` ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v2-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gptj,EleutherAI/gpt-j-6b,EleutherAI/gpt-j-6b,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v2-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gptj,EleutherAI/gpt-j-6b,EleutherAI/gpt-j-6b,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -15224,7 +15224,7 @@ ChildProcessError: Traceback (most recent call last): torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 68.00 MiB. GPU 0 has a total capacity of 14.74 GiB of which 46.12 MiB is free. Process 56491 has 14.69 GiB memory in use. Of the allocated memory 14.29 GiB is allocated by PyTorch, and 313.73 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v2-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neo,EleutherAI/gpt-neo-125m,EleutherAI/gpt-neo-125m,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v2-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neo,EleutherAI/gpt-neo-125m,EleutherAI/gpt-neo-125m,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -15357,7 +15357,7 @@ ChildProcessError: Traceback (most recent call last): AttributeError: 'EmissionsTracker' object has no attribute '_scheduler' ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v2-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neo,EleutherAI/gpt-neo-1.3B,EleutherAI/gpt-neo-1.3B,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v2-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neo,EleutherAI/gpt-neo-1.3B,EleutherAI/gpt-neo-1.3B,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -15570,7 +15570,7 @@ Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 4bit-gptq-exllama-v2-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-1.3b,EleutherAI/pythia-1.3b,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.223-212.873.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.4.0,,4.44.2,,0.34.0,,,,1.21.4,,,,0.12.0,,,MB,1582.342144,1596.915712,0.0,1201.668096,1189.151232,s,1,8.6505029296875,8.6505029296875,0.0,8.6505029296875,8.6505029296875,8.6505029296875,8.6505029296875,[8.6505029296875],,kWh,3.685549087500135e-05,4.0580436646186965e-06,1.1927509542004194e-05,5.284104408162424e-05,,MB,1563.049984,1794.048,0.0,1384.12032,1351.367168,s,10,0.4121473541259766,0.04121473541259766,0.0002022758176152118,0.04114734268188476,0.04153757705688477,0.041589892578124996,0.04163174499511719,"[0.04164220809936523, 0.04098281478881836, 0.041086112976074216, 0.04125983810424805, 0.04104713439941406, 0.041223487854003905, 0.04108512115478516, 0.04152595138549805, 0.04118703842163086, 0.04110764694213867]",tokens/s,6211.370701211666,kWh,1.2486447791132654e-06,1.3770349550795996e-07,8.256404990683729e-07,2.211988773689598e-06,tokens/kWh,115732956.26315133,MB,1572.442112,1835.99104,0.0,1426.06336,1407.548416,s,10,10.61557373046875,1.061557373046875,0.005552198334380235,1.0607288208007812,1.0706404907226563,1.0711919494628908,1.071633116455078,"[1.0594141845703124, 1.0570184326171874, 1.0705179443359376, 1.06061181640625, 1.0608458251953126, 1.0575704345703125, 1.071743408203125, 1.0526453857421876, 1.062847900390625, 1.0623583984375]",tokens/s,59.34676881304852,kWh,3.098585049172095e-05,3.4171700681130072e-06,1.5428455837531704e-05,4.983147639736567e-05,tokens/kWh,1264261.1568966173,,s,630,10.611997209548951,0.016844440015157062,0.0002947726225189542,0.01678551959991455,0.017162866401672364,0.01731639394760132,0.017957598571777342,"[0.017451072692871095, 0.017297407150268555, 0.017342464447021484, 0.01744895935058594, 0.01720684814453125, 0.017137151718139648, 0.017303680419921873, 0.017027904510498047, 0.01723353576660156, 0.0169168643951416, 0.016577983856201173, 0.016558048248291015, 0.016688800811767577, 0.016571327209472655, 0.016566272735595702, 0.01682009506225586, 0.016686399459838866, 0.016845632553100585, 0.016680959701538087, 0.016692256927490233, 0.016493024826049803, 0.016673023223876954, 0.016505088806152344, 0.01654560089111328, 0.016789695739746095, 0.016633440017700195, 0.016837024688720705, 0.016939008712768554, 0.016987327575683595, 0.016929599761962892, 0.016754079818725585, 0.016622047424316406, 0.016738431930541992, 0.0167587833404541, 0.016670719146728515, 0.016709632873535156, 0.016766592025756834, 0.016744384765625, 0.016711679458618164, 0.016721887588500975, 0.01676131248474121, 0.016676000595092774, 0.016565088272094727, 0.016502784729003905, 0.016496639251708984, 0.01643712043762207, 0.017168512344360353, 0.016590368270874022, 0.017137247085571287, 0.016565120697021485, 0.016664287567138673, 0.016777503967285157, 0.016833536148071288, 0.016686080932617187, 0.016695295333862305, 0.016498239517211914, 0.016601280212402345, 0.016535423278808595, 0.01654207992553711, 0.0186079044342041, 0.01702729606628418, 0.01662156867980957, 0.016537599563598633, 0.016503583908081054, 0.016575647354125977, 0.016554847717285156, 0.016551008224487306, 0.01651395225524902, 0.016324607849121094, 0.016479583740234376, 0.016461471557617187, 0.01647702407836914, 0.01651113510131836, 0.016631296157836914, 0.01672809600830078, 0.01696201515197754, 0.01697996711730957, 0.016798944473266603, 0.01676144027709961, 0.0167524471282959, 0.016743904113769532, 0.01680476760864258, 0.016920576095581053, 0.01657151985168457, 0.01662656021118164, 0.016687103271484375, 0.016846847534179688, 0.016939008712768554, 0.01695884895324707, 0.01686355209350586, 0.016666431427001954, 0.016719808578491212, 0.016720447540283203, 0.01716223907470703, 0.016729663848876954, 0.01656262397766113, 0.016525312423706053, 0.016654144287109374, 0.016635391235351564, 0.016847551345825194, 0.01699737548828125, 0.016933696746826172, 0.016909631729125976, 0.016720703125, 0.016834592819213866, 0.016748575210571288, 0.017229824066162108, 0.016768896102905273, 0.016633663177490234, 0.016742719650268554, 0.016685056686401366, 0.016654016494750977, 0.016572351455688476, 0.01679782485961914, 0.016810239791870116, 0.016915840148925783, 0.016792192459106445, 0.017161920547485353, 0.016748287200927733, 0.016734207153320312, 0.016648767471313476, 0.016643104553222657, 0.016561119079589844, 0.017476640701293945, 0.017296255111694334, 0.01794451141357422, 0.017136831283569336, 0.01717945671081543, 0.017006591796875, 0.017069536209106444, 0.016922399520874022, 0.016787967681884765, 0.016718080520629883, 0.016755807876586915, 0.01679657554626465, 0.016773120880126953, 0.016802976608276367, 0.01673097610473633, 0.016815935134887695, 0.01702112007141113, 0.016847936630249024, 0.01666099166870117, 0.016709407806396483, 0.016681631088256835, 0.016644096374511717, 0.018257568359375, 0.019118431091308594, 0.01719478416442871, 0.017072351455688476, 0.017295360565185547, 0.017141632080078125, 0.016875232696533203, 0.01685696029663086, 0.01693503952026367, 0.01691484832763672, 0.016858848571777343, 0.016818464279174803, 0.01671500778198242, 0.016831167221069337, 0.01713145637512207, 0.018671743392944334, 0.01787017631530762, 0.017562271118164063, 0.017112640380859374, 0.016931264877319337, 0.01722127914428711, 0.016879968643188477, 0.01690345573425293, 0.016857376098632814, 0.016716224670410156, 0.01686028861999512, 0.016845695495605467, 0.016742399215698242, 0.016725215911865234, 0.016838464736938476, 0.016762943267822267, 0.016718751907348634, 0.01678335952758789, 0.016811391830444337, 0.016959104537963867, 0.016870399475097657, 0.016831935882568358, 0.016806463241577148, 0.016734207153320312, 0.016711679458618164, 0.016660255432128908, 0.017047775268554687, 0.016889312744140624, 0.01686787223815918, 0.017353471755981446, 0.017210752487182616, 0.01717737579345703, 0.016963136672973632, 0.016852800369262694, 0.016831104278564452, 0.01692803192138672, 0.01665865516662598, 0.01672857666015625, 0.016793344497680666, 0.01693712043762207, 0.016992223739624022, 0.016904319763183594, 0.016969280242919924, 0.017099199295043947, 0.017080320358276366, 0.016892927169799805, 0.016773183822631835, 0.016613471984863282, 0.016587615966796875, 0.016647647857666016, 0.016623584747314454, 0.016553728103637696, 0.016537919998168945, 0.016626176834106447, 0.016721920013427736, 0.016703487396240235, 0.017117183685302736, 0.016946367263793945, 0.016955839157104493, 0.016781696319580076, 0.016842111587524414, 0.01670003128051758, 0.016695295333862305, 0.01661270332336426, 0.016765600204467775, 0.01696710395812988, 0.016869951248168945, 0.016885759353637696, 0.016709632873535156, 0.0167106876373291, 0.016724960327148437, 0.01666662406921387, 0.01658399963378906, 0.016628416061401367, 0.01665433692932129, 0.01660518455505371, 0.017032224655151366, 0.0168253116607666, 0.0169671688079834, 0.016806047439575197, 0.016836959838867186, 0.01687321662902832, 0.0172359676361084, 0.016883935928344727, 0.01679145622253418, 0.01678335952758789, 0.01681011199951172, 0.016683008193969725, 0.01677516746520996, 0.016932159423828124, 0.016926496505737305, 0.016823200225830077, 0.01686387252807617, 0.016883712768554687, 0.01677052879333496, 0.016775711059570313, 0.016723968505859374, 0.01682431983947754, 0.01679974365234375, 0.016901920318603516, 0.01700864028930664, 0.01696995162963867, 0.01699827194213867, 0.017452928543090822, 0.01718092727661133, 0.016957439422607423, 0.016793279647827147, 0.016757055282592772, 0.016690303802490234, 0.01674025535583496, 0.016690143585205076, 0.016697343826293946, 0.016676864624023437, 0.017032480239868163, 0.017193695068359376, 0.01687049674987793, 0.016724063873291017, 0.016806720733642578, 0.01682431983947754, 0.01680998420715332, 0.017360895156860352, 0.017012224197387696, 0.016830976486206056, 0.016864831924438477, 0.01696988868713379, 0.01690140724182129, 0.016843679428100587, 0.017008447647094728, 0.01709699249267578, 0.017082368850708008, 0.0168407039642334, 0.016713727951049806, 0.016695295333862305, 0.016752639770507814, 0.01679897689819336, 0.0168721923828125, 0.0169881591796875, 0.01683865547180176, 0.016904191970825197, 0.016815488815307617, 0.016633823394775392, 0.016732831954956055, 0.01656831932067871, 0.01681203269958496, 0.016695295333862305, 0.01658060836791992, 0.01659596824645996, 0.016638528823852538, 0.01656262397766113, 0.016623615264892578, 0.016632863998413086, 0.016630752563476563, 0.017035039901733398, 0.01665660858154297, 0.01661337661743164, 0.01661337661743164, 0.01660518455505371, 0.016677984237670897, 0.016710559844970704, 0.016738304138183592, 0.01687084770202637, 0.01699897575378418, 0.016915935516357422, 0.016595487594604493, 0.017386816024780274, 0.01676767921447754, 0.01663711929321289, 0.01654457664489746, 0.01667647933959961, 0.017084096908569334, 0.017472192764282225, 0.01714995193481445, 0.016780799865722656, 0.016638463973999023, 0.016817663192749025, 0.01654550361633301, 0.016712480545043946, 0.01670252799987793, 0.016652416229248047, 0.016610111236572266, 0.01677926445007324, 0.016563552856445313, 0.01661404800415039, 0.01765171241760254, 0.016590848922729492, 0.016531455993652345, 0.016547840118408205, 0.016600128173828124, 0.0165283203125, 0.016530847549438475, 0.01652387237548828, 0.01656831932067871, 0.016573728561401366, 0.016622304916381836, 0.01652236747741699, 0.01661427116394043, 0.016621183395385743, 0.016637632369995117, 0.016906335830688478, 0.017150367736816406, 0.017219776153564452, 0.01697750473022461, 0.016859136581420898, 0.01674870491027832, 0.01677337646484375, 0.016680608749389647, 0.016644447326660157, 0.016742399215698242, 0.01666662406921387, 0.017180543899536133, 0.017081663131713866, 0.016775999069213867, 0.016821504592895508, 0.016838464736938476, 0.01674336051940918, 0.01681407928466797, 0.01729030418395996, 0.01687443161010742, 0.017922048568725587, 0.017962944030761718, 0.017193023681640624, 0.01722572708129883, 0.01738137626647949, 0.016969728469848632, 0.016934911727905275, 0.018257568359375, 0.017338720321655274, 0.018175199508666993, 0.01744156837463379, 0.017542463302612304, 0.017328832626342775, 0.017270784378051757, 0.01712656021118164, 0.017015199661254882, 0.016795904159545898, 0.016736064910888672, 0.017590175628662108, 0.01686895942687988, 0.01683340835571289, 0.01686275291442871, 0.01694748878479004, 0.016906015396118163, 0.0169005126953125, 0.01682156753540039, 0.017078975677490234, 0.017127424240112304, 0.01694476890563965, 0.016777599334716797, 0.01681782341003418, 0.0167541446685791, 0.01681702423095703, 0.01677926445007324, 0.016979135513305665, 0.016814943313598632, 0.01694921684265137, 0.017285120010375975, 0.017466367721557616, 0.017293664932250978, 0.017140384674072265, 0.016879615783691407, 0.01703321647644043, 0.017006591796875, 0.017008319854736328, 0.016900415420532226, 0.016743743896484375, 0.017015487670898437, 0.016602975845336914, 0.01669264030456543, 0.016739072799682616, 0.01672985649108887, 0.016734464645385742, 0.016562175750732423, 0.01662156867980957, 0.016611328125, 0.016620800018310546, 0.016610048294067384, 0.016682815551757813, 0.016639999389648438, 0.016533695220947265, 0.016504512786865235, 0.016662847518920897, 0.01666489601135254, 0.01659516716003418, 0.016832735061645506, 0.016553567886352538, 0.01653398323059082, 0.016618623733520507, 0.016832832336425782, 0.016665151596069336, 0.01658470344543457, 0.016664575576782227, 0.01684480094909668, 0.016701440811157226, 0.01678303909301758, 0.01663167953491211, 0.016662975311279297, 0.016668479919433595, 0.0165762882232666, 0.01668943977355957, 0.016667999267578126, 0.01648259162902832, 0.01657244873046875, 0.016601343154907227, 0.01662518310546875, 0.016603839874267577, 0.016564224243164064, 0.01657651138305664, 0.016479711532592773, 0.0165949764251709, 0.016429567337036134, 0.0165150089263916, 0.016563743591308595, 0.016546239852905275, 0.01658684730529785, 0.016486400604248046, 0.016728063583374024, 0.01696476745605469, 0.016734880447387697, 0.01664224052429199, 0.016639999389648438, 0.016734207153320312, 0.01668035125732422, 0.01697443199157715, 0.017004480361938478, 0.01674553680419922, 0.016666976928710938, 0.016740447998046876, 0.016773183822631835, 0.01672652816772461, 0.016781312942504883, 0.016699392318725585, 0.016760223388671874, 0.016706144332885742, 0.01675468826293945, 0.016752479553222656, 0.016826528549194336, 0.01679088020324707, 0.016910560607910158, 0.016717920303344725, 0.016873823165893555, 0.016855039596557618, 0.01680499267578125, 0.01682111930847168, 0.017536735534667967, 0.017137567520141603, 0.01711680030822754, 0.01692310333251953, 0.01679132843017578, 0.01684707260131836, 0.016873151779174804, 0.016877887725830078, 0.016928543090820314, 0.016996383666992188, 0.01693244743347168, 0.016800352096557617, 0.016756032943725584, 0.016765056610107423, 0.01681056022644043, 0.016893951416015626, 0.017053152084350588, 0.01720944023132324, 0.017316287994384765, 0.017229663848876954, 0.017062047958374023, 0.016963584899902344, 0.016957439422607423, 0.01697737693786621, 0.016949792861938477, 0.017065568923950194, 0.01710483169555664, 0.016903999328613282, 0.01678767967224121, 0.01681657600402832, 0.016982015609741212, 0.0170700798034668, 0.016920576095581053, 0.016910335540771485, 0.016903392791748045, 0.01680668830871582, 0.01684480094909668, 0.017215360641479494, 0.01670297622680664, 0.01661734390258789, 0.016546560287475587, 0.016672479629516603, 0.016572160720825194, 0.016679328918457033, 0.016694528579711914, 0.016683839797973634, 0.0167445125579834, 0.016752639770507814, 0.016517120361328123, 0.017156095504760743, 0.01731328010559082, 0.01665279960632324, 0.016615423202514648, 0.016730112075805666, 0.016581823348999023, 0.016572799682617188, 0.017783231735229492, 0.0166146240234375, 0.01653619194030762, 0.016703647613525392, 0.016515071868896485, 0.01661248016357422, 0.016515104293823243, 0.016530271530151366, 0.01681135940551758, 0.016821920394897463, 0.016870399475097657, 0.01670512008666992, 0.016639968872070313, 0.01674630355834961, 0.016577152252197264, 0.016578367233276367, 0.016578752517700194, 0.016664447784423827, 0.017078399658203125, 0.016770368576049806, 0.016636608123779296, 0.016602239608764648, 0.016633823394775392, 0.016505760192871095, 0.01659235191345215, 0.016730655670166017, 0.016494272232055664, 0.016560447692871093, 0.016543359756469728, 0.01651545524597168, 0.016695295333862305, 0.016926528930664063, 0.016590816497802734, 0.01664841651916504, 0.016757984161376954, 0.01670822334289551, 0.016474271774291994, 0.0167891845703125, 0.017135263442993164, 0.016675071716308595, 0.01675817680358887, 0.016991231918334963, 0.01683046340942383, 0.017059648513793945, 0.01690812873840332, 0.017029087066650392, 0.016902336120605467, 0.016955583572387696, 0.016957439422607423, 0.017147903442382813, 0.016701440811157226, 0.01692188835144043, 0.016795520782470704, 0.01686409568786621, 0.017022335052490234, 0.01731648063659668, 0.0171615047454834, 0.01712611198425293, 0.017074176788330078, 0.017657632827758788, 0.016963808059692383, 0.016977920532226562, 0.016920576095581053, 0.016924671173095703, 0.016932384490966797, 0.01713404846191406, 0.017041120529174805, 0.017170719146728516, 0.017114591598510743, 0.017281120300292968, 0.017107168197631837]",tokens/s,59.36677022804998,, 4bit-gptq-exllama-v2-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-1.4b,EleutherAI/pythia-1.4b,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.223-212.873.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.4.0,,4.44.2,,0.34.0,,,,1.21.4,,,,0.12.0,,,MB,1583.456256,1596.915712,0.0,1201.668096,1189.151232,s,1,8.8428955078125,8.8428955078125,0.0,8.8428955078125,8.8428955078125,8.8428955078125,8.8428955078125,[8.8428955078125],,kWh,3.590696299166514e-05,3.950185491363577e-06,1.1906676192000726e-05,5.176382467502944e-05,,MB,1568.436224,1794.048,0.0,1384.12032,1351.367168,s,10,0.4130821113586426,0.04130821113586426,0.00022286802072661455,0.0412576961517334,0.041507010269165034,0.04170185794830322,0.041857736091613765,"[0.041896705627441404, 0.04109718322753906, 0.04121372985839844, 0.04126678466796875, 0.04124860763549805, 0.04128707122802734, 0.041103679656982424, 0.04146371078491211, 0.04115500640869141, 0.0413496322631836]",tokens/s,6197.315084838856,kWh,1.2452644686701731e-06,1.3733006921265444e-07,8.294391978297931e-07,2.2120337357126205e-06,tokens/kWh,115730603.86329415,MB,1573.920768,1835.99104,0.0,1426.06336,1407.548416,s,10,10.498130493164062,1.0498130493164062,0.006266048482070716,1.0488837890625,1.0573603149414061,1.0606197937011719,1.0632273767089844,"[1.056635986328125, 1.0469267578125, 1.04680224609375, 1.0506383056640625, 1.044119140625, 1.0502557373046875, 1.0475118408203126, 1.040291748046875, 1.0510694580078126, 1.0638792724609376]",tokens/s,60.01068479861527,kWh,3.0614132719246836e-05,3.376261120950881e-06,1.530290704136997e-05,4.929330088156768e-05,tokens/kWh,1278064.1359637105,,s,630,10.494630895614618,0.01665814427875337,0.00033934837523612196,0.016566128730773927,0.016881436538696292,0.0172128849029541,0.018070274982452398,"[0.017091936111450195, 0.016477983474731447, 0.016464927673339843, 0.017317983627319337, 0.01666227149963379, 0.016607168197631837, 0.01690451240539551, 0.016965375900268555, 0.01659507179260254, 0.016795520782470704, 0.016619136810302734, 0.01656412887573242, 0.016537471771240233, 0.016523359298706054, 0.016473920822143554, 0.01665273666381836, 0.016604799270629883, 0.0167325439453125, 0.01671833610534668, 0.016717664718627928, 0.01679315185546875, 0.016848735809326172, 0.01670729637145996, 0.016689952850341798, 0.016637184143066405, 0.016526079177856444, 0.016634016036987304, 0.0166910400390625, 0.016626943588256837, 0.016745088577270507, 0.016730239868164062, 0.01675823974609375, 0.0175164794921875, 0.02031407928466797, 0.016833152770996094, 0.017135200500488282, 0.016765344619750978, 0.016703487396240235, 0.016674943923950195, 0.01668492889404297, 0.016623231887817384, 0.016543231964111327, 0.016556768417358397, 0.016605344772338868, 0.016571584701538085, 0.016638784408569335, 0.016842975616455078, 0.017002368927001955, 0.016908191680908204, 0.016742080688476563, 0.016611648559570313, 0.016763904571533202, 0.01654854393005371, 0.016836095809936523, 0.01661836814880371, 0.01681952095031738, 0.01661372756958008, 0.016633216857910155, 0.01659587287902832, 0.01660313606262207, 0.01661452865600586, 0.01654431915283203, 0.01653094482421875, 0.016677728652954103, 0.016698400497436525, 0.016706335067749024, 0.01661894416809082, 0.016435327529907225, 0.01647385597229004, 0.016509408950805663, 0.016431583404541015, 0.016676095962524413, 0.016424959182739257, 0.016535167694091798, 0.01664080047607422, 0.016694623947143553, 0.016685600280761718, 0.016835071563720702, 0.016699392318725585, 0.016648000717163085, 0.01658060836791992, 0.01659903907775879, 0.016561216354370117, 0.01668611145019531, 0.016582784652709962, 0.01675388717651367, 0.016906816482543944, 0.016691328048706055, 0.016530912399291994, 0.01665065574645996, 0.016477695465087892, 0.016539840698242186, 0.016596895217895508, 0.01665043258666992, 0.016730112075805666, 0.01648371124267578, 0.016382816314697266, 0.016558080673217773, 0.016523263931274415, 0.0165515193939209, 0.016525728225708008, 0.016504608154296874, 0.01641878318786621, 0.01638015937805176, 0.01647830390930176, 0.01655388832092285, 0.016680959701538087, 0.016590848922729492, 0.016530527114868163, 0.016690080642700195, 0.01649228858947754, 0.01688118362426758, 0.016558847427368163, 0.0170383358001709, 0.016556480407714843, 0.01659721565246582, 0.01640790367126465, 0.01655833625793457, 0.017318592071533204, 0.01661136054992676, 0.01659699249267578, 0.01671196746826172, 0.01649839973449707, 0.01676268768310547, 0.016574655532836914, 0.0167587833404541, 0.016698400497436525, 0.01652214431762695, 0.01661756706237793, 0.016695327758789062, 0.016475839614868162, 0.016708608627319335, 0.01663692855834961, 0.016599231719970704, 0.016619327545166016, 0.01657206344604492, 0.0166376953125, 0.017625696182250978, 0.016693248748779296, 0.0167093448638916, 0.016582944869995116, 0.016862720489501954, 0.016586528778076173, 0.01645654487609863, 0.016459648132324218, 0.016478208541870116, 0.016449535369873047, 0.01668908882141113, 0.0165, 0.016587392807006836, 0.016574111938476563, 0.016785184860229493, 0.01666223907470703, 0.01662054443359375, 0.016556032180786134, 0.016719871520996094, 0.016582656860351562, 0.016476320266723632, 0.016555871963500977, 0.016547840118408205, 0.016521215438842773, 0.016586143493652342, 0.0167491512298584, 0.016582399368286132, 0.0165295352935791, 0.016610687255859374, 0.016523935317993163, 0.018104415893554687, 0.01661747169494629, 0.01643123245239258, 0.016514080047607422, 0.01648441505432129, 0.016457984924316407, 0.0164234561920166, 0.016478208541870116, 0.016489791870117187, 0.01632736015319824, 0.016474367141723633, 0.01647385597229004, 0.017044479370117188, 0.016540735244750977, 0.01654368019104004, 0.016496639251708984, 0.016483840942382814, 0.0164564151763916, 0.01658412742614746, 0.0165479679107666, 0.01646636772155762, 0.016494367599487306, 0.01680588722229004, 0.01665238380432129, 0.016529312133789064, 0.016668832778930665, 0.016694496154785157, 0.016510719299316405, 0.01649158477783203, 0.016447231292724608, 0.016648384094238283, 0.016777088165283203, 0.017125280380249023, 0.017897247314453125, 0.016877887725830078, 0.016652288436889647, 0.01661337661743164, 0.01656345558166504, 0.016556800842285155, 0.01660927963256836, 0.016559648513793945, 0.016801279067993165, 0.0179866886138916, 0.016797536849975585, 0.017172479629516603, 0.0168222713470459, 0.016748544692993163, 0.01660927963256836, 0.016494783401489257, 0.016852800369262694, 0.016554048538208008, 0.017024959564208984, 0.01648182487487793, 0.016572895050048827, 0.01644361686706543, 0.016498207092285156, 0.016445695877075197, 0.01670070457458496, 0.016646879196166992, 0.016883712768554687, 0.016496448516845702, 0.016517311096191405, 0.016487808227539064, 0.016517536163330078, 0.016473600387573242, 0.016496864318847657, 0.016605695724487304, 0.01653286361694336, 0.016490144729614256, 0.01662393569946289, 0.01665705680847168, 0.016703487396240235, 0.016768768310546876, 0.01654537582397461, 0.016490144729614256, 0.01661849594116211, 0.01662774467468262, 0.01667068862915039, 0.016600896835327148, 0.016529024124145506, 0.01638243293762207, 0.016512287139892577, 0.016646528244018556, 0.016551488876342772, 0.016448095321655275, 0.01669647979736328, 0.016658912658691405, 0.01663363265991211, 0.01656425666809082, 0.016624191284179687, 0.016760831832885743, 0.016678848266601563, 0.016529632568359376, 0.016622623443603515, 0.016550336837768555, 0.016636287689208985, 0.01677654457092285, 0.01663657569885254, 0.016647647857666016, 0.017009279251098634, 0.01671721649169922, 0.016531103134155272, 0.016533504486083983, 0.016513919830322264, 0.016727455139160158, 0.016565984725952148, 0.016540512084960938, 0.016497695922851562, 0.0165644474029541, 0.016497407913208008, 0.016523008346557618, 0.016498943328857422, 0.016519264221191408, 0.016525087356567384, 0.0165295352935791, 0.016733407974243164, 0.016671232223510742, 0.01678531265258789, 0.01650636863708496, 0.016445728302001954, 0.016415327072143555, 0.016467967987060548, 0.01642291259765625, 0.016472063064575194, 0.01648748779296875, 0.01635568046569824, 0.01649135971069336, 0.016602880477905275, 0.016811168670654297, 0.016786272048950196, 0.016473472595214842, 0.016496992111206053, 0.016427295684814453, 0.016483583450317384, 0.01646227264404297, 0.01642527961730957, 0.01653932762145996, 0.016414751052856447, 0.016462112426757814, 0.01643519973754883, 0.016469215393066405, 0.016443487167358398, 0.016548288345336913, 0.01655628776550293, 0.016645824432373047, 0.01662393569946289, 0.01670908737182617, 0.0165053768157959, 0.016819839477539063, 0.016869184494018554, 0.016781280517578125, 0.016736799240112305, 0.016488191604614257, 0.016576927185058595, 0.016425312042236326, 0.016528032302856446, 0.016493408203125, 0.016719104766845704, 0.01693503952026367, 0.016679328918457033, 0.016483871459960938, 0.01656028747558594, 0.016478912353515625, 0.01673539161682129, 0.01669206428527832, 0.016670719146728515, 0.016525312423706053, 0.016612991333007813, 0.01650931167602539, 0.016554208755493165, 0.016469280242919923, 0.016458240509033203, 0.016526975631713868, 0.016889984130859376, 0.016519071578979493, 0.01675254440307617, 0.016612831115722655, 0.016677888870239257, 0.016588544845581053, 0.016454879760742187, 0.016422048568725586, 0.016567583084106444, 0.016529983520507812, 0.016545024871826172, 0.01655593681335449, 0.01648691177368164, 0.016471616744995116, 0.01655084800720215, 0.016504640579223632, 0.016524896621704102, 0.016412992477416993, 0.016513151168823244, 0.016455039978027344, 0.016406431198120117, 0.01655471992492676, 0.01663145637512207, 0.016683040618896486, 0.016557695388793946, 0.01654035186767578, 0.016527328491210937, 0.016754720687866213, 0.016684255599975585, 0.016522016525268555, 0.016726015090942382, 0.016720096588134767, 0.01678927993774414, 0.016739904403686525, 0.017084863662719725, 0.017033376693725587, 0.01822697639465332, 0.018415679931640627, 0.017523136138916016, 0.017217536926269532, 0.016865280151367186, 0.016986112594604492, 0.01660927963256836, 0.016615423202514648, 0.01656831932067871, 0.016570367813110352, 0.016635168075561525, 0.016456703186035156, 0.016532735824584963, 0.016640480041503907, 0.016758687973022462, 0.016892000198364256, 0.0168175048828125, 0.016799999237060548, 0.016800159454345702, 0.01676313591003418, 0.01666227149963379, 0.016699392318725585, 0.016564224243164064, 0.016740352630615234, 0.016547359466552735, 0.016535072326660155, 0.016573375701904296, 0.016566272735595702, 0.01648454475402832, 0.01657360076904297, 0.016485023498535156, 0.016541919708251952, 0.01646335983276367, 0.016529695510864258, 0.016518367767333984, 0.01638275146484375, 0.016678720474243163, 0.016558080673217773, 0.016748191833496094, 0.016793535232543944, 0.016501344680786133, 0.016424959182739257, 0.01639833641052246, 0.01640457534790039, 0.016539072036743162, 0.016466400146484376, 0.01652889633178711, 0.01662822341918945, 0.016666656494140626, 0.016626976013183595, 0.0167425594329834, 0.01664463996887207, 0.01664521598815918, 0.016526144027709962, 0.016506975173950195, 0.016456800460815428, 0.016733152389526367, 0.016617408752441408, 0.016581920623779296, 0.016483327865600587, 0.016498559951782225, 0.016408063888549804, 0.01655228805541992, 0.016513023376464844, 0.016495967864990236, 0.016802175521850586, 0.01655193519592285, 0.01660745620727539, 0.01645955276489258, 0.016465919494628906, 0.01660927963256836, 0.016547168731689453, 0.016519392013549804, 0.01663337516784668, 0.01652934455871582, 0.016509920120239257, 0.01645484733581543, 0.016497472763061523, 0.016451583862304688, 0.016582271575927735, 0.0165230712890625, 0.016457664489746095, 0.016456319808959962, 0.016445119857788085, 0.01638387107849121, 0.016455423355102538, 0.01646771240234375, 0.01648736000061035, 0.01650499153137207, 0.016417856216430663, 0.016458528518676758, 0.016433120727539063, 0.016455135345458984, 0.016427583694458008, 0.016424959182739257, 0.016465280532836916, 0.016411264419555663, 0.01653913688659668, 0.01661939239501953, 0.0166015682220459, 0.016494752883911133, 0.01639423942565918, 0.0164270076751709, 0.01652118492126465, 0.01658064079284668, 0.016453472137451172, 0.016449695587158204, 0.016459840774536133, 0.01642255973815918, 0.01646620750427246, 0.016565568923950197, 0.016538303375244142, 0.0164454402923584, 0.016476160049438478, 0.016455455780029295, 0.016521087646484377, 0.01648089599609375, 0.016430816650390624, 0.01643734359741211, 0.016543968200683594, 0.01641212844848633, 0.016506399154663086, 0.016597312927246095, 0.016727935791015624, 0.01656268882751465, 0.016418560028076172, 0.0164552001953125, 0.017152448654174805, 0.016714784622192384, 0.01652124786376953, 0.016458688735961916, 0.016565311431884767, 0.01648736000061035, 0.01645523262023926, 0.016470464706420898, 0.016441343307495117, 0.016541696548461913, 0.016490848541259765, 0.016506528854370116, 0.016561983108520507, 0.016670143127441406, 0.016548608779907225, 0.017452959060668946, 0.019414623260498046, 0.01674678421020508, 0.01699567985534668, 0.016751487731933593, 0.016631999969482423, 0.01649203109741211, 0.01672137641906738, 0.016611743927001953, 0.01671340751647949, 0.016554752349853517, 0.016625631332397462, 0.016477407455444334, 0.016512096405029295, 0.01646972846984863, 0.016660480499267577, 0.016531455993652345, 0.016476160049438478, 0.016463872909545898, 0.016461599349975587, 0.016363616943359374, 0.01657644844055176, 0.01645996856689453, 0.016438592910766603, 0.016446144104003906, 0.016442752838134764, 0.01646860885620117, 0.01660518455505371, 0.01653555107116699, 0.016453632354736326, 0.016568191528320314, 0.017970304489135742, 0.01645804786682129, 0.016741056442260743, 0.016471647262573243, 0.01645199966430664, 0.016377504348754884, 0.016375167846679688, 0.016376800537109375, 0.01666975975036621, 0.018975679397583007, 0.016713727951049806, 0.017374656677246095, 0.016542272567749025, 0.01646518325805664, 0.016378591537475586, 0.01664723205566406, 0.01645254325866699, 0.0164270076751709, 0.016817663192749025, 0.016674911499023438, 0.016589664459228517, 0.01655948829650879, 0.016436960220336912, 0.01651299285888672, 0.016752960205078125, 0.016716415405273438, 0.016699392318725585, 0.016584928512573243, 0.01650787162780762, 0.016466720581054688, 0.01647830390930176, 0.01650886344909668, 0.016529632568359376, 0.016445215225219727, 0.01644476890563965, 0.016406784057617186, 0.016482208251953127, 0.016546239852905275, 0.016580671310424806, 0.016555807113647462, 0.016865503311157225, 0.01696998405456543, 0.017135360717773437, 0.01738479995727539, 0.01736969566345215, 0.017295295715332032, 0.01719718360900879, 0.01723391914367676, 0.017320032119750976, 0.017207199096679688, 0.01739366340637207, 0.017253503799438477, 0.017158687591552733, 0.017285472869873048, 0.01745305633544922, 0.017662208557128908, 0.017401599884033205, 0.01873823928833008, 0.016734336853027342, 0.01657734489440918, 0.01651456069946289, 0.016600576400756836, 0.016484640121459962, 0.017732320785522462, 0.01759052848815918, 0.017706111907958986, 0.016631935119628908, 0.01669990348815918, 0.016725887298583986, 0.016681087493896483, 0.016531455993652345, 0.01656012725830078, 0.01653539276123047, 0.016534879684448243, 0.01654662322998047, 0.01658060836791992, 0.016697343826293946, 0.01696713638305664, 0.016904735565185548, 0.01698406410217285, 0.01694633674621582]",tokens/s,60.03069629283077,, -4bit-gptq-exllama-v2-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v2-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -15615,7 +15615,7 @@ ChildProcessError: Traceback (most recent call last): self._buffers[key] = fn(buf) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1160, in convert return t.to( -torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 70.00 MiB. GPU 0 has a total capacity of 14.74 GiB of which 16.12 MiB is free. Process 21442 has 14.72 GiB memory in use. Of the allocated memory 14.44 GiB is allocated by PyTorch, and 187.67 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 70.00 MiB. GPU 0 has a total capacity of 14.74 GiB of which 16.12 MiB is free. Process 21490 has 14.72 GiB memory in use. Of the allocated memory 14.44 GiB is allocated by PyTorch, and 187.67 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 4bit-gptq-exllama-v2-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,gpt_neox,EleutherAI/pythia-12b,EleutherAI/pythia-12b,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.223-212.873.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.4.0,,4.44.2,,0.34.0,,,,1.21.4,,,,0.12.0,,,MB,6768.418816,7762.542592,0.0,7367.294976,7351.94368,s,1,12.4044169921875,12.4044169921875,0.0,12.4044169921875,12.4044169921875,12.4044169921875,12.4044169921875,[12.4044169921875],,kWh,0.00015872177171666712,1.7501032018332932e-05,5.042504033999848e-05,0.00022664784407499852,,MB,1417.826304,8404.271104,0.0,7994.343424,7863.794176,s,10,2.982906372070313,0.29829063720703125,0.00042369488986336985,0.29844981384277347,0.2986946716308594,0.298739990234375,0.2987762451171875,"[0.29835421752929686, 0.298471435546875, 0.29781222534179685, 0.2973498840332031, 0.2985828857421875, 0.2979638671875, 0.2987853088378906, 0.2984737548828125, 0.29868460083007814, 0.2984281921386719]",tokens/s,858.2233837340357,kWh,8.727604025857624e-06,9.621517366482535e-07,5.779473577823583e-06,1.546922934032946e-05,tokens/kWh,16548982.135301886,MB,1438.72,8676.900864,0.0,8266.973184,8120.408064,s,10,24.584218749999998,2.458421875,0.0029808773235690616,2.459900634765625,2.4606286865234375,2.4616724975585935,2.462507546386719,"[2.452980712890625, 2.454088134765625, 2.455698974609375, 2.45816455078125, 2.460396728515625, 2.45989404296875, 2.4599072265625, 2.460341064453125, 2.460031005859375, 2.46271630859375]",tokens/s,25.626195666681497,kWh,7.195771981330942e-05,7.937285538448352e-06,4.791437493277619e-05,0.00012780938028453392,tokens/kWh,492921.56694404664,,s,630,24.580845401763934,0.03901721492343479,0.000344952758432959,0.03898783874511719,0.0394716007232666,0.03956559257507324,0.03973260978698731,"[0.03861248016357422, 0.03829539108276367, 0.03821443176269531, 0.03829779052734375, 0.03813715362548828, 0.03831273651123047, 0.03855257415771484, 0.03852585601806641, 0.038358047485351564, 0.03833673477172852, 0.03850726318359375, 0.03876169586181641, 0.038717952728271485, 0.03879759979248047, 0.03879935836791992, 0.03861094284057617, 0.03860070419311523, 0.03877068710327149, 0.03867407989501953, 0.03870896148681641, 0.038890113830566404, 0.038834175109863284, 0.038780033111572264, 0.038701313018798825, 0.038986366271972654, 0.0389918098449707, 0.03897932815551758, 0.03889593505859375, 0.0387665901184082, 0.0386448974609375, 0.038695743560791016, 0.03877020645141602, 0.03877289581298828, 0.038809951782226564, 0.038801185607910155, 0.0389073600769043, 0.03910294342041016, 0.03914115142822266, 0.03913356781005859, 0.039319679260253905, 0.03939328002929687, 0.039329792022705076, 0.0392674560546875, 0.039203712463378906, 0.0392086067199707, 0.039346527099609375, 0.03924790573120117, 0.03914070510864258, 0.03931523132324219, 0.03948531341552734, 0.0394024658203125, 0.039255840301513675, 0.03922723388671875, 0.03942643356323242, 0.03933171081542969, 0.03926796722412109, 0.039246337890625, 0.0392171516418457, 0.039272449493408204, 0.03922323226928711, 0.03929884719848633, 0.03952041625976563, 0.03955513763427734, 0.038827072143554686, 0.03851753616333008, 0.03855091094970703, 0.038583072662353515, 0.03852492904663086, 0.03839299011230469, 0.038617759704589846, 0.038627231597900394, 0.038451297760009766, 0.03856150436401367, 0.03861516952514649, 0.03841878509521485, 0.0384532470703125, 0.03838908767700195, 0.03837020874023438, 0.038631168365478516, 0.03880550384521484, 0.03878870391845703, 0.03872604751586914, 0.038714752197265626, 0.03894745635986328, 0.0390041618347168, 0.03880361557006836, 0.03894255828857422, 0.03891404724121094, 0.0388403205871582, 0.0389119987487793, 0.038801406860351564, 0.03885055923461914, 0.03902582550048828, 0.039082817077636715, 0.03902211380004883, 0.03895347213745117, 0.038891166687011716, 0.03887926483154297, 0.03924720001220703, 0.03915673446655273, 0.038903423309326175, 0.03881961441040039, 0.038906143188476565, 0.039137569427490235, 0.03899987030029297, 0.039051456451416014, 0.039292160034179686, 0.03930393600463867, 0.03919257736206055, 0.03913667297363281, 0.03911945724487305, 0.039049217224121094, 0.03936870574951172, 0.03941904067993164, 0.03942627334594727, 0.03941798400878906, 0.039391742706298825, 0.039346176147460936, 0.03925196838378906, 0.03952435302734375, 0.03949318313598633, 0.039307712554931644, 0.03919462585449219, 0.03927859115600586, 0.039299072265625, 0.03924905776977539, 0.038626625061035154, 0.038763198852539066, 0.03858163070678711, 0.03854985427856445, 0.03858233642578125, 0.038663585662841796, 0.03869779205322266, 0.03858169555664062, 0.038343231201171876, 0.03857779312133789, 0.038680320739746095, 0.03871603012084961, 0.03865190505981445, 0.03864371109008789, 0.03881132888793945, 0.03872998428344727, 0.03859462356567383, 0.0385986557006836, 0.03855155181884766, 0.03850649642944336, 0.03859443283081055, 0.038905982971191404, 0.038817790985107424, 0.038778881072998046, 0.038978912353515624, 0.038981441497802735, 0.0388043212890625, 0.038760448455810545, 0.03880755233764648, 0.03896713638305664, 0.038965408325195315, 0.03894668960571289, 0.03908607864379883, 0.03936249542236328, 0.03899571228027344, 0.03882438278198242, 0.038895584106445315, 0.03898076629638672, 0.03914012908935547, 0.03916400146484375, 0.039182464599609376, 0.0390695686340332, 0.0390203857421875, 0.03896950531005859, 0.03929292678833008, 0.03948086547851563, 0.03934163284301758, 0.03939420700073242, 0.03938508987426758, 0.039346176147460936, 0.039354366302490236, 0.039376449584960935, 0.039239425659179684, 0.039158462524414066, 0.03902668762207031, 0.039139102935791016, 0.03956694412231445, 0.03963881683349609, 0.039430335998535154, 0.0395015983581543, 0.03937750244140625, 0.03935043334960937, 0.03955904006958008, 0.03853311920166016, 0.03847577667236328, 0.03842863845825195, 0.03855791854858399, 0.03859151840209961, 0.03849929428100586, 0.03844281768798828, 0.03857619094848633, 0.03845113754272461, 0.038703102111816406, 0.038694911956787106, 0.03857161712646484, 0.0385577278137207, 0.038689216613769534, 0.038739902496337894, 0.03869900894165039, 0.03885465621948242, 0.03862720108032226, 0.038645313262939456, 0.03885113525390625, 0.03873532867431641, 0.038713886260986326, 0.03894796752929688, 0.03891811370849609, 0.03890473556518555, 0.03889152145385742, 0.03890991973876953, 0.03883625411987305, 0.039034881591796876, 0.03887513732910156, 0.03882393646240234, 0.03878092956542969, 0.03893657684326172, 0.039204864501953124, 0.039041023254394534, 0.039051136016845706, 0.03925619125366211, 0.03938051223754883, 0.039223041534423825, 0.03907043075561523, 0.03951001739501953, 0.03932950210571289, 0.03937516784667969, 0.039411582946777345, 0.039241825103759766, 0.03940156936645508, 0.039444385528564455, 0.039241825103759766, 0.03925187301635742, 0.03925763320922852, 0.0390865592956543, 0.03913318252563477, 0.03952230453491211, 0.039354366302490236, 0.03937257766723633, 0.039276737213134766, 0.03941584014892578, 0.039534591674804685, 0.03957356643676758, 0.03943008041381836, 0.039632190704345704, 0.0396479377746582, 0.03960543823242187, 0.039279361724853516, 0.038668449401855466, 0.038520862579345706, 0.03839740753173828, 0.03834726333618164, 0.038634719848632815, 0.0387305908203125, 0.03861004638671875, 0.038561729431152346, 0.03854415893554688, 0.038735904693603516, 0.0391550407409668, 0.03881817626953125, 0.03862563323974609, 0.03878092956542969, 0.039002113342285157, 0.03884236907958984, 0.03881564712524414, 0.03890528106689453, 0.03897529602050781, 0.03894768142700195, 0.03896121597290039, 0.03897235107421875, 0.03884134292602539, 0.03871705627441406, 0.038954910278320314, 0.03894102478027344, 0.038885505676269534, 0.038852222442626955, 0.038873374938964846, 0.0389571533203125, 0.03891167831420898, 0.039134719848632815, 0.039058238983154296, 0.03891814422607422, 0.039099681854248045, 0.03903948974609375, 0.03905110549926758, 0.039325759887695315, 0.03927219009399414, 0.03918700790405273, 0.03935612869262695, 0.03938742446899414, 0.03935356903076172, 0.03924614334106445, 0.0391602897644043, 0.03917168045043945, 0.03911721420288086, 0.03921100616455078, 0.03916595077514649, 0.0393994255065918, 0.0394439697265625, 0.03930777740478516, 0.03929235076904297, 0.03925459289550781, 0.03939123153686523, 0.039462913513183595, 0.039376895904541014, 0.03956531143188476, 0.039702529907226565, 0.03959807968139648, 0.039669761657714846, 0.03957715225219727, 0.03886896133422851, 0.03856921768188477, 0.03834969711303711, 0.03845951843261719, 0.03861830520629883, 0.03855817413330078, 0.03854275131225586, 0.03844112014770508, 0.038855457305908205, 0.03879267120361328, 0.0388265266418457, 0.03873791885375977, 0.03864950561523438, 0.03856118392944336, 0.03865103912353516, 0.03884009552001953, 0.03874816131591797, 0.03859008026123047, 0.038805343627929687, 0.03889539337158203, 0.038808319091796876, 0.03893779373168945, 0.03915449523925781, 0.03909222412109375, 0.038950912475585936, 0.03888742446899414, 0.0389172477722168, 0.03879328155517578, 0.038770561218261716, 0.03885782241821289, 0.03895280075073242, 0.0388485107421875, 0.03905507278442383, 0.03899820709228516, 0.03895036697387695, 0.03907551956176758, 0.039140289306640624, 0.03908198547363281, 0.039139102935791016, 0.039233470916748045, 0.03922358322143555, 0.0392171516418457, 0.03928799819946289, 0.03935702514648438, 0.039321823120117186, 0.03929087829589844, 0.03918745422363281, 0.039330814361572264, 0.039327743530273435, 0.039317089080810545, 0.03942031860351562, 0.03946905517578125, 0.03938889694213867, 0.03929116821289062, 0.03955712127685547, 0.03947510528564453, 0.039497825622558595, 0.03927449417114258, 0.03978035354614258, 0.039702529907226565, 0.03962166213989258, 0.03956582260131836, 0.03964156723022461, 0.0387740478515625, 0.03865468978881836, 0.03848515319824219, 0.0384903678894043, 0.03848259353637695, 0.03833139038085937, 0.03831084823608399, 0.03867238235473633, 0.038963359832763673, 0.03888470458984375, 0.03877529525756836, 0.03855974578857422, 0.03846259307861328, 0.03853612899780273, 0.03865593719482422, 0.03863532638549805, 0.03879539108276367, 0.03892025756835937, 0.038899711608886715, 0.03879935836791992, 0.03883827209472656, 0.03911884689331055, 0.03889583969116211, 0.038849857330322264, 0.03883055877685547, 0.039013504028320316, 0.03883712005615234, 0.038983680725097655, 0.03895296096801758, 0.038829345703125, 0.038793952941894534, 0.03890176010131836, 0.03916799926757813, 0.0391044807434082, 0.03905235290527344, 0.03948992156982422, 0.039219806671142575, 0.03920444869995117, 0.03913532638549805, 0.03897993469238281, 0.039061473846435546, 0.039256065368652344, 0.03911065673828125, 0.03908403015136719, 0.03925571060180664, 0.03928940963745117, 0.03920054244995117, 0.03917737579345703, 0.03952931213378906, 0.039411712646484375, 0.03940505599975586, 0.03939718246459961, 0.03936735916137695, 0.03938304138183594, 0.039396991729736326, 0.039857887268066404, 0.03961718368530273, 0.03955913543701172, 0.039511199951171874, 0.03949606323242188, 0.039352832794189455, 0.03987251281738281, 0.039744895935058595, 0.03879305648803711, 0.038383838653564456, 0.03863929748535156, 0.038570240020751954, 0.03851446533203125, 0.03897366333007812, 0.038870849609375, 0.038674625396728515, 0.03870851135253906, 0.038645633697509764, 0.03857084655761719, 0.038662143707275394, 0.038746208190917966, 0.03861222457885742, 0.038547199249267576, 0.0387694091796875, 0.03891238403320312, 0.03880527877807617, 0.03866419219970703, 0.03882572937011719, 0.03885696029663086, 0.03892351913452148, 0.03892844772338867, 0.03887142562866211, 0.038959423065185544, 0.03874745559692383, 0.03873452758789062, 0.03881369781494141, 0.039008255004882815, 0.03901030349731445, 0.03892428970336914, 0.03927859115600586, 0.039142494201660154, 0.03909856033325195, 0.03904608154296875, 0.03909609603881836, 0.03922681427001953, 0.03950678253173828, 0.039323360443115234, 0.03938508987426758, 0.039090175628662106, 0.03908198547363281, 0.03912499237060547, 0.03924991989135742, 0.03923164749145508, 0.039182079315185546, 0.03915164947509766, 0.03938515090942383, 0.039362560272216796, 0.039479297637939455, 0.039446529388427735, 0.039384735107421874, 0.039368030548095706, 0.03925484848022461, 0.03967830276489258, 0.039626590728759764, 0.03948339080810547, 0.03938620758056641, 0.03933638381958008, 0.03926883316040039, 0.03943971252441406, 0.03948716735839844, 0.03966432189941406, 0.038912448883056644, 0.03842047882080078, 0.03872742462158203, 0.03859072113037109, 0.038569854736328124, 0.03854355239868164, 0.03854713439941406, 0.03867212677001953, 0.03849615859985352, 0.038480480194091796, 0.03863740921020508, 0.03857424163818359, 0.038563838958740236, 0.03868057632446289, 0.03870230484008789, 0.038683425903320315, 0.038965248107910154, 0.038860321044921875, 0.03895548629760742, 0.038983680725097655, 0.03871088027954102, 0.03858451080322266, 0.038713569641113284, 0.03868880081176758, 0.03864476776123047, 0.03899488067626953, 0.03907788848876953, 0.038948863983154294, 0.03892838287353516, 0.03907788848876953, 0.03898931121826172, 0.03939583969116211, 0.039403518676757815, 0.03932364654541016, 0.03915776062011719, 0.039203937530517576, 0.03899417495727539, 0.038847137451171875, 0.038978816986083985, 0.03901708984375, 0.039118785858154294, 0.039233409881591796, 0.03934828948974609, 0.03916825485229492, 0.03950732803344727, 0.0393570556640625, 0.03930316925048828, 0.039634944915771485, 0.03928473663330078, 0.039137279510498044, 0.03946080017089844, 0.039446590423583984, 0.03945209503173828, 0.03939395141601563, 0.039411617279052735, 0.03942399978637695, 0.039577598571777346, 0.039585792541503906, 0.03958169555664062, 0.03951599884033203, 0.03954643249511719, 0.03947148895263672, 0.039472606658935545, 0.038924415588378905, 0.03868239974975586, 0.03857827377319336, 0.038459007263183596, 0.03850243377685547, 0.03862972640991211, 0.03869510269165039, 0.038467391967773434, 0.038682174682617185, 0.03880799865722656, 0.038703102111816406, 0.0388403205871582, 0.03902668762207031, 0.0388485107421875, 0.03883404922485351, 0.038856510162353516, 0.03890950393676758, 0.038855422973632814, 0.03883747100830078, 0.038722335815429686, 0.03912086486816406, 0.03922332763671875, 0.0388935661315918, 0.038982719421386716, 0.0388474235534668, 0.038808895111083985, 0.03883078384399414, 0.038787071228027346, 0.03878297424316406, 0.038973438262939454, 0.03902054214477539, 0.03886899185180664, 0.039419902801513675, 0.039411712646484375, 0.03919257736206055, 0.03927014541625977, 0.03912054443359375, 0.03913584136962891, 0.0389911994934082, 0.03899868774414063, 0.03917391967773438, 0.03915711975097656, 0.03905593490600586, 0.03911231994628906, 0.03919529724121094, 0.03926220703125, 0.03926806259155274, 0.039200958251953126, 0.03953468704223633, 0.0394048957824707, 0.039430816650390624, 0.039419902801513675, 0.039282047271728515, 0.0397973747253418, 0.03960319900512695, 0.03955542373657227, 0.03957827377319336, 0.03954035186767578, 0.03997119903564453, 0.039751201629638674, 0.03964483261108399, 0.03944736099243164, 0.039436286926269534]",tokens/s,25.629712473387563,, @@ -15699,7 +15699,7 @@ ValueError: XGLMForCausalLM does not support an attention implementation through raise RuntimeError(f""Isolated process exited with non-zero code {isolated_process.exitcode}"") RuntimeError: Isolated process exited with non-zero code -9 ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-gptq-exllama-v2-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,Deci/DeciCoder-1b,Deci/DeciCoder-1b,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.759616,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): +4bit-gptq-exllama-v2-sdpa,pytorch,2.4.1+cu124,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,llama,Deci/DeciCoder-1b,Deci/DeciCoder-1b,cuda,0,42,,,True,True,,float16,True,False,,sdpa,,False,,False,forward,gptq,4,True,2,256,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz,8,33163.751424,Linux,x86_64,Linux-5.10.225-213.878.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['Tesla T4'],1,16106127360,0.5.0,,4.45.1,,0.34.2,,,,1.22.0,,,,0.13.0,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 153, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch