experiment,dataset,successful_calls,errors,acceptance_length,acceptance_rate_pct,position_1_pct,position_2_pct,position_3_pct,position_4_pct,position_5_pct,position_6_pct,position_7_pct,position_8_pct BASIC4 T0,HumanEval,6113,0,4.807433568954425,47.59291961193031,83.48910457231432,68.2955461615601,56.05040420414842,46.438411701067665,39.082074736092274,33.35858251014397,28.844422877439253,25.184810132676535 BASIC4 T0,GSM8K,112565,0,5.8452949929301266,60.566187411626586,89.111124649917,78.43314942568261,69.42534995217603,61.56513870444342,54.91736490150955,48.928548868838995,43.53965352202276,38.60916926842232 BASIC4 T0,MATH500,17996,0,5.315660395956892,53.945754949461154,86.4338302641223,73.8404691747072,63.20606152624376,54.37304319945573,47.034425086389675,40.77491824898309,35.334843618094816,30.568448477692662 BASIC4 T0,MT-Bench,4875,0,3.6381420740851347,32.976775926064185,74.9640303021086,54.310188295953424,39.48269998092062,29.28122358727341,22.363451100105905,17.565171746666408,14.180632863906885,11.666809531578215 SPEED-Bench T0,coding,89,0,4.0971,38.71,78.33,60.15,46.26,36.22,28.99,23.66,19.59,16.51 SPEED-Bench T0,humanities,88,0,3.8826,36.03,76.53,57.73,43.58,33.12,25.87,20.46,16.82,14.16 SPEED-Bench T0,math,98,0,3.9028,36.28,77.84,59.14,44.81,34.08,26.16,20.21,15.73,12.3 SPEED-Bench T0,qa,80,0,4.2508,40.64,80.65,63.71,49.73,38.84,30.85,24.68,20.06,16.56 SPEED-Bench T0,rag,105,0,4.9694,49.62,84.57,70.38,58.39,48.39,41.03,35.35,31.23,27.59 SPEED-Bench T0,reasoning,190,0,4.5232,44.04,81.24,65.24,51.8,41.39,34.17,29.37,25.84,23.27 SPEED-Bench T0,stem,86,0,3.8932,36.17,77.98,59.32,44.71,33.55,25.57,19.78,15.73,12.67 SPEED-Bench T0,writing,84,0,3.1636,27.04,70.81,48.53,32.9,22.57,15.76,11.23,8.27,6.27 SPEED-Bench T0,multilingual,80,0,3.9722,37.15,75.38,57.57,44.4,35.04,28.12,22.69,18.64,15.38 SPEED-Bench T0,summarization,80,0,4.629,45.36,85.01,69.37,55.91,45.3,36.43,28.88,23.32,18.67 SPEED-Bench T0,roleplay,158,0,2.9572,24.47,67.65,43.17,27.69,18.57,13.21,10.11,8.3,7.01 BASIC4 T1,humaneval,164,0,4.294067058567386,41.17583823209232,78.17929457870673,61.04670150228609,48.516220335292836,39.07794469845417,32.30677117352493,27.15708687132593,23.161332462442846,19.961354234704984 BASIC4 T1,gsm8k,1319,0,4.662285405179695,45.77856756474619,79.8866882408207,64.5829576836571,53.301130270562666,44.72777656515426,37.97841951923168,32.70572157953176,28.346919988991488,24.698926670019834 BASIC4 T1,math500,500,0,4.632882069795427,45.41102587244284,80.61321700762134,65.35148415563577,53.61311672683514,44.62845968712395,37.59025270758123,31.784997994384277,26.94494584837545,22.76173285198556 BASIC4 T1,mtbench,160,0,3.095874052030335,26.19842565037919,67.74426418354612,45.31208601324757,30.91177882307766,21.625803974272824,15.704329461457233,11.828741480272631,9.20840933090141,7.251991936258039 SPEED-Bench T1,coding,89,0,3.7522630712699994,34.40328839087499,73.83975732986298,54.35868483503024,40.78217101737732,31.3069356196892,24.800486105794995,19.984402510908673,16.45774324427773,13.696126464058793 SPEED-Bench T1,humanities,78,10,2.8185892204995118,22.732365256243895,63.61483820047356,40.73883322408465,26.604506396621762,17.83451277317744,12.423804618766694,8.951034295167641,6.6441034696489325,5.0472890720104875 SPEED-Bench T1,math,88,10,3.481256654324715,31.01570817905894,72.07726167585692,51.40952502026093,37.66519940462002,27.85493704545334,21.03031458733917,16.058044250928294,12.416638857548454,9.613744590464385 SPEED-Bench T1,multilingual,80,0,3.498545566544608,31.231819581807603,69.5424523459156,50.07699941822662,37.05383114883132,28.233120016426543,21.867834776359466,17.454912562882857,14.140515382772664,11.484891003045755 SPEED-Bench T1,qa,79,1,3.1040785289644686,26.300981612055857,69.17185123738422,47.02889534079912,32.15816397068989,22.065532973869765,15.285497027512788,10.996820129959906,7.902668325729296,5.798423890501867 SPEED-Bench T1,rag,105,0,3.8383407546285446,35.47925943285681,72.8521209280525,54.408249355519104,42.06233887977502,33.016170611670965,26.833841106163582,21.837356456526834,17.956409655495666,14.867588469650808 SPEED-Bench T1,reasoning,189,1,2.8580765924236786,23.225957405295986,63.283560249007685,40.626543341605135,26.90917072635372,18.538390717695766,13.239965612062605,9.87189883605368,7.515349580211083,5.822780179378213 SPEED-Bench T1,roleplay,158,0,2.4504565641942078,18.130707052427596,61.07780819025645,35.241779205404875,20.091840658719207,11.655016927937506,7.141402945234917,4.555839572918316,3.0832221141448186,2.198746804804669 SPEED-Bench T1,stem,78,8,3.002715318869165,25.033941485864563,66.85820731974579,44.85163269778654,30.34845496383958,20.8531667762437,14.379793995178611,10.174008327854482,7.408941485864562,5.397326320403243 SPEED-Bench T1,summarization,80,0,4.29741136747327,41.21764209341587,81.46689176514725,64.72519227161884,51.15362971299944,40.20821609454136,31.523166385293568,24.826486587882197,19.864940911648848,15.97261301819546 SPEED-Bench T1,writing,80,4,2.772608882265594,22.15761102831992,64.7225991517087,41.07388025283453,26.214386828291662,16.991685039202146,11.295443700064249,7.687406526671056,5.4181873438778805,3.8572993839091247 MRCR T0,8 needles,458,342,6.207113290053785,65.08891612567231,92.19037613698606,84.60747604809238,75.13221812173944,65.37950417081953,57.71399581571173,52.20124091982509,48.23020768795625,45.25631010424804