Add V100 benchmarking results

data/2023-06-17/V100_chat-concise_benchmark.csv

@@ -0,0 +1,21 @@
+model,throughput,response_length,latency,energy
+camel-ai/CAMEL-13B-Combined-Data,21.43101835743948,280.880120886501,13.064448516246319,2699.068023841508
+lmsys/fastchat-t5-3b-v1.0,17.265044888298704,310.34318334452655,23.454799682692173,1043.990802216248
+lmsys/vicuna-13B,21.221113733316532,268.2877770315648,12.474878561076359,2412.714352921421
+lmsys/vicuna-7B,33.19220689261821,273.57118871725993,8.18447234641069,1387.838511417057
+metaai/llama-7B,23.92946132846408,68.0584284754869,2.2809002959847486,354.6560946944264
+project-baize/baize-v2-7B,30.350461249705614,313.3747481531229,10.034955017859746,1647.9252511752854
+BAIR/koala-13b,21.426431872159036,253.86501007387508,11.723909422593445,2576.3193623237044
+BlinkDL/RWKV-4-Raven-7B-v12-Eng98%-Other2%-20230521-ctx8192.pth,36.40036863259138,202.84654130288786,4.8940369672449,1139.9740564136987
+FreedomIntelligence/phoenix-inst-chat-7b,42.268276726494044,233.5211551376763,5.210845160281831,1237.6637434519814
+Neutralzz/BiLLa-7B-SFT,31.9073779709284,141.02820685023505,4.304014573938894,791.1183122901291
+StabilityAI/stablelm-tuned-alpha-7b,27.433760323389556,241.67763599731364,8.396188638064837,1539.8807639355257
+metaai/llama-13B,17.16666100714585,85.41638683680323,4.213923745980638,936.364457689723
+openaccess-ai-collective/manticore-13b-chat-pyg,21.501265310987147,274.15983881799866,12.659875248785934,2867.592066151776
+tatsu-lab/alpaca-7B,30.27856487351534,124.94392209536602,3.96404224254347,697.6676591672265
+BAIR/koala-7b,31.656500750351455,247.8448623237072,7.693811395496254,1251.9351081262555
+OpenAssistant/oasst-sft-1-pythia-12b,15.208750214131749,249.4513096037609,15.906977624923046,2461.867333445271
+databricks/dolly-v2-12b,14.77422282257357,145.3566151779718,9.38322044956342,1392.614104096707
+h2oai/h2ogpt-gm-oasst1-en-2048-open-llama-7b-preview-300bt-v2,31.09362733624043,213.68334452652786,6.7498229851656,1084.1301658831444
+nomic-ai/gpt4all-13b-snoozy,21.536649161834035,212.530557421088,9.801092169845262,1977.1679271323053
+togethercomputer/RedPajama-INCITE-7B-Chat,21.663817805785996,273.98455339153793,12.047950960830523,1659.1448901947638

data/2023-06-17/V100_chat_benchmark.csv

@@ -0,0 +1,21 @@
+model,throughput,response_length,latency,energy
+BAIR/koala-13b,21.533160265051638,265.5540631296172,12.202378193294793,2299.2539251175217
+BlinkDL/RWKV-4-Raven-7B-v12-Eng98%-Other2%-20230521-ctx8192.pth,38.25693186890057,214.35460040295501,5.09708168833379,1112.947044996645
+Neutralzz/BiLLa-7B-SFT,31.761691850702793,160.1222296843519,4.935750165560669,783.9250990597718
+StabilityAI/stablelm-tuned-alpha-7b,28.53967935283584,242.85829415715244,8.213054464106607,1349.8225080590983
+h2oai/h2ogpt-gm-oasst1-en-2048-open-llama-7b-preview-300bt-v2,30.275817945748678,216.17797179314977,6.902688020302885,1083.431902955002
+metaai/llama-13B,17.273409248575422,76.56447280053727,3.7837350834271555,772.3963885157816
+metaai/llama-7B,26.078374550887073,65.01040967092008,2.106037566082945,335.8834066487579
+openaccess-ai-collective/manticore-13b-chat-pyg,21.385993003768778,293.6336467427804,13.589340266513748,2717.3955362659517
+camel-ai/CAMEL-13B-Combined-Data,21.443262786761146,292.39724647414374,13.597516823282726,3063.6723606447304
+lmsys/fastchat-t5-3b-v1.0,17.285271119159383,312.98119543317665,24.00358150668783,1150.395621558095
+lmsys/vicuna-13B,21.71511101442676,278.87676292813967,12.8324168655423,2980.6725436534525
+project-baize/baize-v2-7B,30.553878516266057,324.9533243787777,10.401131271597677,1837.7422095365996
+tatsu-lab/alpaca-7B,30.30927551611021,130.43619879113498,4.1233307988468715,727.2185275352584
+BAIR/koala-7b,32.54393862793764,261.6477501678979,7.92655260342141,1308.97298321021
+FreedomIntelligence/phoenix-inst-chat-7b,43.255453591642365,241.45936870382806,5.358263290288085,1212.8509496306262
+OpenAssistant/oasst-sft-1-pythia-12b,15.360072232285452,256.2988582941571,16.199311907878233,2513.5866148421774
+databricks/dolly-v2-12b,14.8711288386801,151.2726662189389,9.739419852054695,1458.1875147750186
+lmsys/vicuna-7B,32.13252691172293,283.11618535930154,8.759656009372382,1458.5311544660867
+nomic-ai/gpt4all-13b-snoozy,21.59345917467568,244.30624580255204,11.242197763509727,2265.4151793149776
+togethercomputer/RedPajama-INCITE-7B-Chat,22.218321996015874,278.4751511081263,11.953991510078367,1661.3254805238432

data/2023-06-17/V100_instruct-concise_benchmark.csv

@@ -0,0 +1,21 @@
+model,throughput,response_length,latency,energy
+BAIR/koala-7b,31.870929688850776,191.7407656145064,5.776933217154402,940.6980705171267
+BlinkDL/RWKV-4-Raven-7B-v12-Eng98%-Other2%-20230521-ctx8192.pth,39.221009432980566,226.7229684351914,5.369737004086241,1264.4471417058417
+FreedomIntelligence/phoenix-inst-chat-7b,42.470186422244424,193.91034251175284,4.339222575622636,968.785210208196
+lmsys/fastchat-t5-3b-v1.0,32.731683961001515,280.61417058428475,13.955027274680319,699.2501370047005
+lmsys/vicuna-13B,21.749876946304724,194.72666218938886,8.894261652518159,2004.231935191404
+metaai/llama-13B,18.570015239967084,117.85896574882472,5.5800629816924205,1151.47489355272
+metaai/llama-7B,26.579538100451405,104.01846877098724,3.3514236870209313,594.7948838146413
+nomic-ai/gpt4all-13b-snoozy,20.983074234487425,124.95735392881129,5.782305658279372,1160.8125960376083
+togethercomputer/RedPajama-INCITE-7B-Chat,22.13375331287594,273.00067159167224,11.777779012989532,1618.0334294828767
+StabilityAI/stablelm-tuned-alpha-7b,29.00482449092475,223.56044325050368,7.41862470058361,1333.2689402283406
+camel-ai/CAMEL-13B-Combined-Data,21.460150389563363,188.2263263935527,8.713883080348563,1952.1352001343216
+databricks/dolly-v2-12b,14.910884049151369,157.10543989254532,10.048765299306664,1523.7667649429138
+lmsys/vicuna-7B,33.00422508114337,218.19845533915378,6.521462410481351,1134.2387068502344
+openaccess-ai-collective/manticore-13b-chat-pyg,21.69738931527289,229.17763599731364,10.522901259287245,2290.821816655476
+BAIR/koala-13b,20.97945489282466,168.6877098723976,7.777735522072535,1462.6801752854299
+Neutralzz/BiLLa-7B-SFT,30.656448666055088,99.96474143720617,3.1034504787446604,493.75409200805854
+OpenAssistant/oasst-sft-1-pythia-12b,15.311911635208784,239.28374748153124,15.189281238389173,2210.647578240437
+h2oai/h2ogpt-gm-oasst1-en-2048-open-llama-7b-preview-300bt-v2,30.99540256126542,217.50503693754197,6.9492957234086985,1093.5390379449263
+project-baize/baize-v2-7B,30.171092209096468,243.02182672934856,7.826149599316017,1279.5043529214238
+tatsu-lab/alpaca-7B,32.01702487024256,119.6296171927468,3.623346220601726,576.9808673606445

data/2023-06-17/V100_instruct_benchmark.csv

@@ -0,0 +1,21 @@
+model,throughput,response_length,latency,energy
+camel-ai/CAMEL-13B-Combined-Data,21.698576445650616,246.24983210208194,11.359782962929792,2355.2015053727355
+databricks/dolly-v2-12b,15.057831510732873,159.93451981195435,10.197591386884174,1619.948066823369
+nomic-ai/gpt4all-13b-snoozy,21.81850230944136,220.82639355271994,10.106196799825966,2202.807249832109
+project-baize/baize-v2-7B,32.278702324630125,300.21826729348555,9.151365330046675,1557.4543243787796
+FreedomIntelligence/phoenix-inst-chat-7b,43.15794896524618,220.48556077904635,4.912834925525496,1134.9532484889164
+Neutralzz/BiLLa-7B-SFT,31.576098085051683,146.43485560779047,4.467502261238455,821.7476809939546
+OpenAssistant/oasst-sft-1-pythia-12b,15.249688094942522,252.08865010073876,16.127927775413387,2645.1371722632653
+h2oai/h2ogpt-gm-oasst1-en-2048-open-llama-7b-preview-300bt-v2,31.9963132875822,228.91537944929482,7.081518725379781,1231.1418035594343
+lmsys/fastchat-t5-3b-v1.0,30.11501180390337,359.11014103425117,18.113400371066604,872.8616652115497
+BAIR/koala-13b,21.686819170677825,255.0466756212223,11.667351228473493,2196.7527931497716
+StabilityAI/stablelm-tuned-alpha-7b,28.54267912639095,233.76057756883816,7.929687720685966,1381.3434637340479
+lmsys/vicuna-13B,21.92047685400391,264.0950302216253,12.061023608554358,2516.2876131632
+openaccess-ai-collective/manticore-13b-chat-pyg,21.703210621473303,274.1773002014775,12.581106571755186,2362.644396574878
+togethercomputer/RedPajama-INCITE-7B-Chat,21.543639351812168,281.8143049026192,12.532223938875198,1700.600847212891
+lmsys/vicuna-7B,31.797633189426183,268.90094022834114,8.36020701868699,1386.7618619879124
+metaai/llama-13B,18.476088005501783,117.27736736064473,5.550619761657461,1274.2525688381443
+metaai/llama-7B,23.642990385313613,110.3025520483546,3.7306664026356073,636.4017854264604
+tatsu-lab/alpaca-7B,31.257537640928863,127.87004701141706,3.9853428024686144,671.8288287441229
+BAIR/koala-7b,31.414143726094665,253.60376091336468,7.832867446137829,1181.3637347212878
+BlinkDL/RWKV-4-Raven-7B-v12-Eng98%-Other2%-20230521-ctx8192.pth,39.9399840394634,246.79650772330425,5.83791405340994,1302.0306672263273

data/2023-06-17/schema.yaml

@@ -1,2 +1,2 @@
-gpu: ["A100", "A40"]
+gpu: ["A100", "A40", "V100"]
 task: ["chat", "chat-concise", "instruct", "instruct-concise"]