performance-llm-board / data /2024-01-25 13:30:17.207984_summary.csv
piotr-szleg-bards-ai's picture
2024-02-09 09:57:45 Publish script update
8530cb4
model,language,template_name,execution_time,characters_count,words_count,chunk_sizes_mean,chunk_sizes_min,chunk_sizes_max,chunk_generation_times_mean,chunk_generation_times_min,chunk_generation_times_max
Mistral-7B-Instruct-v0.2,english,call,,,,,,,,,
Mistral-7B-Instruct-v0.2,english,json,7.318650853633881,222.2,35.95,4.157156220767072,1.0,14.0,0.10865272596910582,0.00024437904357910156,60.7178213596344
Mistral-7B-Instruct-v0.2,english,markdown,2.8044421672821045,238.15,38.35,4.148954703832753,1.0,14.0,0.03463017442085187,0.020945072174072266,0.25538039207458496
Mistral-7B-Instruct-v0.2,japanese,call,,,,,,,,,
Mistral-7B-Instruct-v0.2,japanese,json,1.929928070306778,37.9,2.95,1.3882783882783882,1.0,11.0,0.039187116718990896,0.020466089248657227,0.23994874954223633
Mistral-7B-Instruct-v0.2,japanese,markdown,1.8897509932518006,28.85,1.15,1.088679245283019,1.0,4.0,0.0396594097029488,0.0004932880401611328,0.2205650806427002
Mixtral-8x7B-Instruct-v0.1,english,call,,,,,,,,,
Mixtral-8x7B-Instruct-v0.1,english,json,3.181837737560272,394.95,61.5,4.355665839536807,1.0,15.0,0.015196637342151676,0.0002071857452392578,1.011472463607788
Mixtral-8x7B-Instruct-v0.1,english,markdown,3.399144798517227,654.4,105.05,4.374331550802139,1.0,14.0,0.013296874289844125,0.00021576881408691406,0.5904912948608398
Mixtral-8x7B-Instruct-v0.1,japanese,call,,,,,,,,,
Mixtral-8x7B-Instruct-v0.1,japanese,json,8.82588157138309,532.425,56.35,2.0712896323672436,1.0,15.0,0.027260187998098102,0.0002186298370361328,76.45081758499146
Mixtral-8x7B-Instruct-v0.1,japanese,markdown,12.566328430175782,460.6,5.7,1.0741604477611941,1.0,9.0,0.01511352992991903,0.00021529197692871094,0.8098442554473877
TinyLlama/TinyLlama-1.1B-Chat-v1.0,english,call,,,,,,,,,
TinyLlama/TinyLlama-1.1B-Chat-v1.0,english,json,1.3998499751091003,203.15,34.6,4.120689655172414,1.0,13.0,0.01180358711652775,0.0003421306610107422,0.13397955894470215
TinyLlama/TinyLlama-1.1B-Chat-v1.0,english,markdown,1.4005955815315247,179.5,30.65,3.847802786709539,1.0,14.0,0.01265179331535184,0.0003273487091064453,0.13492584228515625
TinyLlama/TinyLlama-1.1B-Chat-v1.0,japanese,call,,,,,,,,,
TinyLlama/TinyLlama-1.1B-Chat-v1.0,japanese,json,1.2260903775691987,48.35,5.8,1.95748987854251,1.0,13.0,0.01474765992840292,0.00041413307189941406,0.18952107429504395
TinyLlama/TinyLlama-1.1B-Chat-v1.0,japanese,markdown,1.1986963748931885,51.55,6.85,2.0415841584158416,1.0,10.0,0.014000740617808729,0.0005300045013427734,0.16827392578125
chat-bison,english,call,,,,,,,,,
chat-bison,english,json,3.446989095211029,550.05,86.0,89.4390243902439,1.0,138.0,0.4258628240445765,0.0006544589996337891,2.2015340328216553
chat-bison,japanese,call,,,,,,,,,
chat-bison,japanese,json,4.333498167991638,246.1,12.0,33.48299319727891,3.0,124.0,0.47770044917152044,0.0005426406860351562,3.5425846576690674
chat-bison-32k,english,call,,,,,,,,,
chat-bison-32k,english,json,3.74513863325119,538.4,84.25,88.26229508196721,1.0,134.0,0.47832588094179745,0.0006556510925292969,3.0114917755126953
chat-bison-32k,japanese,call,,,,,,,,,
chat-bison-32k,japanese,json,4.5112169027328495,241.45,11.8,35.77037037037037,6.0,124.0,0.5467217816246881,0.0005869865417480469,4.387756824493408
gemini-pro,english,call,,,,,,,,,
gemini-pro,english,json,3.6837018529574075,557.475,87.225,175.58267716535434,4.0,988.0,1.0232121174729716,0.0006723403930664062,8.669384241104126
gemini-pro,english,markdown,4.207570552825928,663.8,109.05,139.74736842105264,19.0,472.0,0.7348335742950439,0.0007510185241699219,2.7830429077148438
gemini-pro,japanese,call,,,,,,,,,
gemini-pro,japanese,json,4.794884290013995,189.55,4.05,64.2542372881356,4.0,215.0,1.1180692850533178,0.0006062984466552734,5.698127031326294
gemini-pro,japanese,markdown,7.311563323525822,392.7,7.75,62.832,15.0,203.0,0.9826514282226563,0.0006639957427978516,13.754565954208374
gpt-3.5-turbo,english,call,3.409479808062315,671.15625,108.59375,,,,,,
gpt-3.5-turbo,english,json,2.650409721559094,564.6451612903226,90.98387096774194,5.037122302158274,1.0,16.0,0.017705622920029456,0.0005590915679931641,0.4925534725189209
gpt-3.5-turbo,english,markdown,3.1277945737044015,649.3333333333334,103.75,5.186023294509152,1.0,16.0,0.019346605402459322,0.0006084442138671875,1.127073049545288
gpt-3.5-turbo,japanese,call,8.916680586142618,375.1875,4.9375,,,,,,
gpt-3.5-turbo,japanese,json,6.141976299932447,244.75806451612902,3.306451612903226,1.1217474866942638,1.0,7.0,0.0232258041112258,0.0005545616149902344,13.75982117652893
gpt-3.5-turbo,japanese,markdown,8.25130424045381,315.5,1.3636363636363635,1.1153784348385023,1.0,10.0,0.025426305415027686,0.0005831718444824219,13.71114993095398
gpt-4,english,call,10.59899697303772,538.45,83.7,,,,,,
gpt-4,english,json,9.713180017471313,558.35,87.5,4.971950133570792,1.0,16.0,0.07619813660479824,0.0005440711975097656,1.9534873962402344
gpt-4,japanese,call,22.77469485684445,242.1,1.2,,,,,,
gpt-4,japanese,json,19.565327330639487,217.85,4.1,1.1505149194613151,1.0,7.0,0.09166746716213503,0.0005655288696289062,2.1393778324127197
gpt-4-turbo,english,call,9.345100498199463,550.9,84.9,,,,,,
gpt-4-turbo,english,json,8.325499892234802,686.55,105.4,5.042600073448402,1.0,17.0,0.05410054322774035,0.0007143020629882812,0.755082368850708
gpt-4-turbo,japanese,call,18.703535676002502,222.35,3.6,,,,,,
gpt-4-turbo,japanese,json,19.659980058670044,272.5,6.0,1.181956191715463,1.0,8.0,0.07962745783567377,0.0006952285766601562,13.252992153167725
llama-2-70b-chat,english,call,,,,,,,,,
llama-2-70b-chat,english,json,3.55666184425354,427.6363636363636,65.81818181818181,4.172062084257206,1.0,14.0,0.01772472985303587,0.00021791458129882812,0.5100884437561035
llama-2-70b-chat,english,markdown,6.183407843112946,1169.7727272727273,184.1818181818182,4.07263807564488,1.0,16.0,0.01361775794423896,0.0002124309539794922,0.5661592483520508
llama-2-70b-chat,japanese,call,,,,,,,,,
llama-2-70b-chat,japanese,json,4.945841180650811,350.4761904761905,42.642857142857146,2.46483590087073,1.0,15.0,0.01905980879878806,0.00021982192993164062,0.35228896141052246
llama-2-70b-chat,japanese,markdown,14.294062587949965,1299.1818181818182,172.5,2.8476636445152934,1.0,16.0,0.02320206142747451,0.0002117156982421875,76.44498753547668
zephyr-7b-beta,english,call,,,,,,,,,
zephyr-7b-beta,english,json,2.845248854160309,238.1,39.2,4.274685816876122,1.0,14.0,0.03622233246986485,0.0003070831298828125,0.4896988868713379
zephyr-7b-beta,english,markdown,7.618257820606232,238.65,38.75,4.358904109589041,1.0,15.0,0.1235110485390441,0.00040459632873535156,75.15552616119385
zephyr-7b-beta,japanese,call,,,,,,,,,
zephyr-7b-beta,japanese,json,1.921021181344986,29.75,3.3,1.1829025844930416,1.0,7.0,0.042037644869764565,0.018090486526489258,0.4599947929382324
zephyr-7b-beta,japanese,markdown,1.8884475827217102,26.6,1.75,1.0350194552529184,1.0,4.0,0.04097196471366437,0.02583909034729004,0.4606599807739258