You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
swarms/tests/aop/test_data/aop_benchmark_data/benchmark_results.csv

56 KiB

1agent_counttest_namemodel_namelatency_msthroughput_rpsmemory_usage_mbcpu_usage_percentsuccess_rateerror_counttotal_requestsconcurrent_requeststimestampcost_usdtokens_usedresponse_quality_scoreadditional_metricsagent_creation_timetool_registration_timeexecution_timetotal_latencychaining_stepschaining_successerror_scenarios_testedrecovery_rateresource_cyclesavg_memory_deltamemory_leak_detected
21scaling_testgpt-4o-mini1131.70633316040044.1314292246305761.250.01.002051759345643.94532660.0015359999999999996102400.8548663728748707{'min_latency_ms': 562.7951622009277, 'max_latency_ms': 1780.4391384124756, 'p95_latency_ms': np.float64(1744.0685987472534), 'p99_latency_ms': np.float64(1773.1650304794312), 'total_time_s': 4.84093976020813, 'initial_memory_mb': 291.5546875, 'final_memory_mb': 292.8046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 7.679999999999998e-05, 'quality_std': 0.0675424923987846, 'data_size_processed': 1000, 'model_provider': 'gpt'}0.00.00.00.00False00.000.0False
36scaling_testgpt-4o-mini1175.69503784179693.75758540048262770.00.01.002051759345654.2251950.0015359999999999996102400.8563524483655013{'min_latency_ms': 535.4223251342773, 'max_latency_ms': 1985.3930473327637, 'p95_latency_ms': np.float64(1975.6355285644531), 'p99_latency_ms': np.float64(1983.4415435791016), 'total_time_s': 5.322566986083984, 'initial_memory_mb': 293.1796875, 'final_memory_mb': 293.1796875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 7.679999999999998e-05, 'quality_std': 0.05770982402152013, 'data_size_processed': 1000, 'model_provider': 'gpt'}0.00.00.00.00False00.000.0False
411scaling_testgpt-4o-mini996.96847200393684.4960995090291460.00.01.002051759345662.89771990.0015359999999999996102400.8844883644941982{'min_latency_ms': 45.22204399108887, 'max_latency_ms': 1962.2983932495117, 'p95_latency_ms': np.float64(1647.7753758430483), 'p99_latency_ms': np.float64(1899.3937897682185), 'total_time_s': 4.448300123214722, 'initial_memory_mb': 293.5546875, 'final_memory_mb': 293.5546875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 7.679999999999998e-05, 'quality_std': 0.043434832388308614, 'data_size_processed': 1000, 'model_provider': 'gpt'}0.00.00.00.00False00.000.0False
516scaling_testgpt-4o-mini1112.86814212799073.5878339500741270.00.01.002051759345673.1626520.0015359999999999996102400.8563855623109009{'min_latency_ms': 564.1369819641113, 'max_latency_ms': 1951.472282409668, 'p95_latency_ms': np.float64(1897.4883794784546), 'p99_latency_ms': np.float64(1940.6755018234253), 'total_time_s': 5.57439398765564, 'initial_memory_mb': 293.8046875, 'final_memory_mb': 293.8046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 7.679999999999998e-05, 'quality_std': 0.05691925404970228, 'data_size_processed': 1000, 'model_provider': 'gpt'}0.00.00.00.00False00.000.0False
61scaling_testgpt-4o1298.22400808334353.36709955994058460.1250.01.002051759345683.20654250.0512102400.9279627852934385{'min_latency_ms': 693.6078071594238, 'max_latency_ms': 1764.8026943206787, 'p95_latency_ms': np.float64(1681.7602753639221), 'p99_latency_ms': np.float64(1748.1942105293274), 'total_time_s': 5.939830303192139, 'initial_memory_mb': 293.8046875, 'final_memory_mb': 293.9296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00256, 'quality_std': 0.050879141399088765, 'data_size_processed': 1000, 'model_provider': 'gpt'}0.00.00.00.00False00.000.0False
76scaling_testgpt-4o1264.48545455932623.52938261023188460.00.01.002051759345692.64395280.0512102400.9737471278894755{'min_latency_ms': 175.65083503723145, 'max_latency_ms': 1990.2207851409912, 'p95_latency_ms': np.float64(1910.3824019432068), 'p99_latency_ms': np.float64(1974.2531085014343), 'total_time_s': 5.66671347618103, 'initial_memory_mb': 293.9296875, 'final_memory_mb': 293.9296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00256, 'quality_std': 0.038542680129780495, 'data_size_processed': 1000, 'model_provider': 'gpt'}0.00.00.00.00False00.000.0False
811scaling_testgpt-4o1212.06073760986333.7990000043023230.1250.01.002051759345701.87194230.0512102400.9366077507029601{'min_latency_ms': 542.8001880645752, 'max_latency_ms': 1973.801851272583, 'p95_latency_ms': np.float64(1969.2555904388428), 'p99_latency_ms': np.float64(1972.892599105835), 'total_time_s': 5.264543294906616, 'initial_memory_mb': 293.9296875, 'final_memory_mb': 294.0546875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00256, 'quality_std': 0.044670864578792276, 'data_size_processed': 1000, 'model_provider': 'gpt'}0.00.00.00.00False00.000.0False
916scaling_testgpt-4o1367.16319322586063.12297901073146540.00.01.002051759345711.97384430.0512102400.9328922198254587{'min_latency_ms': 715.888261795044, 'max_latency_ms': 1905.6315422058105, 'p95_latency_ms': np.float64(1890.480661392212), 'p99_latency_ms': np.float64(1902.6013660430908), 'total_time_s': 6.404141664505005, 'initial_memory_mb': 294.0546875, 'final_memory_mb': 294.0546875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00256, 'quality_std': 0.05146728864962903, 'data_size_processed': 1000, 'model_provider': 'gpt'}0.00.00.00.00False00.000.0False
101scaling_testgpt-4-turbo1429.13708686828613.31416147440892670.1250.01.002051759345722.76502420.1024102400.960928099222926{'min_latency_ms': 637.6686096191406, 'max_latency_ms': 1994.9300289154053, 'p95_latency_ms': np.float64(1973.6997246742249), 'p99_latency_ms': np.float64(1990.6839680671692), 'total_time_s': 6.0347089767456055, 'initial_memory_mb': 294.0546875, 'final_memory_mb': 294.1796875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00512, 'quality_std': 0.0429193742204114, 'data_size_processed': 1000, 'model_provider': 'gpt'}0.00.00.00.00False00.000.0False
116scaling_testgpt-4-turbo1167.80121326446533.9339465649517240.00.01.002051759345731.8096480.1024102400.9575695597206497{'min_latency_ms': 521.2328433990479, 'max_latency_ms': 1973.503828048706, 'p95_latency_ms': np.float64(1931.3542008399963), 'p99_latency_ms': np.float64(1965.073902606964), 'total_time_s': 5.083953142166138, 'initial_memory_mb': 294.1796875, 'final_memory_mb': 294.1796875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00512, 'quality_std': 0.04742414087184447, 'data_size_processed': 1000, 'model_provider': 'gpt'}0.00.00.00.00False00.000.0False
1211scaling_testgpt-4-turbo1435.19544601440433.07938699531246130.00.01.002051759345741.91177250.1024102400.9564233524947511{'min_latency_ms': 711.4903926849365, 'max_latency_ms': 2034.2109203338623, 'p95_latency_ms': np.float64(1998.979663848877), 'p99_latency_ms': np.float64(2027.1646690368652), 'total_time_s': 6.4947991371154785, 'initial_memory_mb': 294.3046875, 'final_memory_mb': 294.3046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00512, 'quality_std': 0.03428874308764032, 'data_size_processed': 1000, 'model_provider': 'gpt'}0.00.00.00.00False00.000.0False
1316scaling_testgpt-4-turbo1092.10133552551274.0578190532528870.00.01.002051759345749.88339070.1024102400.9521218582720758{'min_latency_ms': 554.4416904449463, 'max_latency_ms': 1968.658447265625, 'p95_latency_ms': np.float64(1637.098050117493), 'p99_latency_ms': np.float64(1902.346367835998), 'total_time_s': 4.92875599861145, 'initial_memory_mb': 294.3046875, 'final_memory_mb': 294.3046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00512, 'quality_std': 0.043763298033728824, 'data_size_processed': 1000, 'model_provider': 'gpt'}0.00.00.00.00False00.000.0False
141scaling_testclaude-3-5-sonnet1046.92368507385254.0474964468760680.00.01.002051759345757.95395180.03071999999999999102400.9511838758969231{'min_latency_ms': 184.94415283203125, 'max_latency_ms': 1966.0136699676514, 'p95_latency_ms': np.float64(1677.8094530105593), 'p99_latency_ms': np.float64(1908.3728265762325), 'total_time_s': 4.941326141357422, 'initial_memory_mb': 294.3046875, 'final_memory_mb': 294.3046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0015359999999999996, 'quality_std': 0.03727295215254124, 'data_size_processed': 1000, 'model_provider': 'claude'}0.00.00.00.00False00.000.0False
156scaling_testclaude-3-5-sonnet1381.37722015380863.2839793432783560.00.01.002051759345768.71533680.03071999999999999102400.957817098536435{'min_latency_ms': 543.0643558502197, 'max_latency_ms': 1937.4654293060303, 'p95_latency_ms': np.float64(1931.4598441123962), 'p99_latency_ms': np.float64(1936.2643122673035), 'total_time_s': 6.090172290802002, 'initial_memory_mb': 294.3046875, 'final_memory_mb': 294.3046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0015359999999999996, 'quality_std': 0.044335695599357156, 'data_size_processed': 1000, 'model_provider': 'claude'}0.00.00.00.00False00.000.0False
1611scaling_testclaude-3-5-sonnet1314.39613103866583.52435214683366560.00.01.002051759345778.62694030.03071999999999999102400.9749641888502683{'min_latency_ms': 535.1722240447998, 'max_latency_ms': 1983.6831092834473, 'p95_latency_ms': np.float64(1918.512487411499), 'p99_latency_ms': np.float64(1970.6489849090576), 'total_time_s': 5.674801826477051, 'initial_memory_mb': 294.3046875, 'final_memory_mb': 294.3046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0015359999999999996, 'quality_std': 0.03856740540886548, 'data_size_processed': 1000, 'model_provider': 'claude'}0.00.00.00.00False00.000.0False
1716scaling_testclaude-3-5-sonnet1120.7208752632143.70280708758075460.00.01.002051759345788.31617020.03071999999999999102400.9344569749738585{'min_latency_ms': 207.9324722290039, 'max_latency_ms': 2018.561601638794, 'p95_latency_ms': np.float64(1963.4979844093323), 'p99_latency_ms': np.float64(2007.5488781929016), 'total_time_s': 5.401307582855225, 'initial_memory_mb': 294.3046875, 'final_memory_mb': 294.3046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0015359999999999996, 'quality_std': 0.04750434388073592, 'data_size_processed': 1000, 'model_provider': 'claude'}0.00.00.00.00False00.000.0False
181scaling_testclaude-3-haiku1268.54013204574583.5399216876522360.00.01.002051759345797.64959050.0256102400.8406194607723803{'min_latency_ms': 534.9514484405518, 'max_latency_ms': 1956.9103717803955, 'p95_latency_ms': np.float64(1938.3319020271301), 'p99_latency_ms': np.float64(1953.1946778297424), 'total_time_s': 5.6498425006866455, 'initial_memory_mb': 294.4296875, 'final_memory_mb': 294.4296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00128, 'quality_std': 0.053962632063170944, 'data_size_processed': 1000, 'model_provider': 'claude'}0.00.00.00.00False00.000.0False
196scaling_testclaude-3-haiku1377.6446938514713.1892122714791640.00.01.002051759345808.21798010.0256102400.8370154862115219{'min_latency_ms': 661.4456176757812, 'max_latency_ms': 2013.9634609222412, 'p95_latency_ms': np.float64(1985.2455973625183), 'p99_latency_ms': np.float64(2008.2198882102966), 'total_time_s': 6.271141052246094, 'initial_memory_mb': 294.4296875, 'final_memory_mb': 294.4296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00128, 'quality_std': 0.057589803133820325, 'data_size_processed': 1000, 'model_provider': 'claude'}0.00.00.00.00False00.000.0False
2011scaling_testclaude-3-haiku1161.99744939804083.67787951328011560.00.01.002051759345817.25412940.0256102400.8421329247896683{'min_latency_ms': 549.6580600738525, 'max_latency_ms': 1785.23588180542, 'p95_latency_ms': np.float64(1730.9520959854126), 'p99_latency_ms': np.float64(1774.3791246414185), 'total_time_s': 5.437916040420532, 'initial_memory_mb': 294.4296875, 'final_memory_mb': 294.4296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00128, 'quality_std': 0.05774508247670216, 'data_size_processed': 1000, 'model_provider': 'claude'}0.00.00.00.00False00.000.0False
2116scaling_testclaude-3-haiku1365.47502279281622.9988214356292510.00.01.002051759345827.87501260.0256102400.8483772503724578{'min_latency_ms': 767.146110534668, 'max_latency_ms': 1936.8767738342285, 'p95_latency_ms': np.float64(1919.3583130836487), 'p99_latency_ms': np.float64(1933.3730816841125), 'total_time_s': 6.669286727905273, 'initial_memory_mb': 294.4296875, 'final_memory_mb': 294.4296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00128, 'quality_std': 0.05705131022796498, 'data_size_processed': 1000, 'model_provider': 'claude'}0.00.00.00.00False00.000.0False
221scaling_testclaude-3-sonnet1360.1875662803653.0895207354500490.00.01.002051759345837.77377270.15360000000000001102400.8835217044830507{'min_latency_ms': 550.3547191619873, 'max_latency_ms': 1977.1480560302734, 'p95_latency_ms': np.float64(1924.659264087677), 'p99_latency_ms': np.float64(1966.6502976417542), 'total_time_s': 6.473495960235596, 'initial_memory_mb': 294.4296875, 'final_memory_mb': 294.4296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.007680000000000001, 'quality_std': 0.058452629496046606, 'data_size_processed': 1000, 'model_provider': 'claude'}0.00.00.00.00False00.000.0False
236scaling_testclaude-3-sonnet1256.1388015747073.47326855640793350.00.01.002051759345848.57010820.15360000000000001102400.8863139635356961{'min_latency_ms': 641.2796974182129, 'max_latency_ms': 1980.7326793670654, 'p95_latency_ms': np.float64(1846.4025855064392), 'p99_latency_ms': np.float64(1953.86666059494), 'total_time_s': 5.758264780044556, 'initial_memory_mb': 294.4296875, 'final_memory_mb': 294.4296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.007680000000000001, 'quality_std': 0.05783521510861833, 'data_size_processed': 1000, 'model_provider': 'claude'}0.00.00.00.00False00.000.0False
2411scaling_testclaude-3-sonnet1306.070089340213.50203473175514950.00.01.002051759345858.64721630.15360000000000001102400.9094961422561505{'min_latency_ms': 591.8083190917969, 'max_latency_ms': 1971.1270332336426, 'p95_latency_ms': np.float64(1944.3620324134827), 'p99_latency_ms': np.float64(1965.7740330696106), 'total_time_s': 5.710965633392334, 'initial_memory_mb': 294.4296875, 'final_memory_mb': 294.4296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.007680000000000001, 'quality_std': 0.042442911768923584, 'data_size_processed': 1000, 'model_provider': 'claude'}0.00.00.00.00False00.000.0False
2516scaling_testclaude-3-sonnet1307.14819431304933.2629388826761320.00.01.002051759345869.9055440.15360000000000001102400.8938240662052681{'min_latency_ms': 646.7251777648926, 'max_latency_ms': 1990.9627437591553, 'p95_latency_ms': np.float64(1935.0676536560059), 'p99_latency_ms': np.float64(1979.7837257385254), 'total_time_s': 6.129443645477295, 'initial_memory_mb': 294.4296875, 'final_memory_mb': 294.4296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.007680000000000001, 'quality_std': 0.04247877605865338, 'data_size_processed': 1000, 'model_provider': 'claude'}0.00.00.00.00False00.000.0False
261scaling_testgemini-1.5-pro1401.34763717651372.9432184905211410.00.01.002051759345881.2382180.0128102400.9409363720199192{'min_latency_ms': 520.9827423095703, 'max_latency_ms': 1970.2589511871338, 'p95_latency_ms': np.float64(1958.1118822097778), 'p99_latency_ms': np.float64(1967.8295373916626), 'total_time_s': 6.7952821254730225, 'initial_memory_mb': 294.4296875, 'final_memory_mb': 294.4296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00064, 'quality_std': 0.05267230653872383, 'data_size_processed': 1000, 'model_provider': 'gemini'}0.00.00.00.00False00.000.0False
276scaling_testgemini-1.5-pro1341.4858341217043.39829515821790240.00.01.002051759345889.55534670.0128102400.9355344625586725{'min_latency_ms': 503.9515495300293, 'max_latency_ms': 1978.0657291412354, 'p95_latency_ms': np.float64(1966.320013999939), 'p99_latency_ms': np.float64(1975.716586112976), 'total_time_s': 5.885303974151611, 'initial_memory_mb': 294.4296875, 'final_memory_mb': 294.4296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00064, 'quality_std': 0.054780000845711954, 'data_size_processed': 1000, 'model_provider': 'gemini'}0.00.00.00.00False00.000.0False
2811scaling_testgemini-1.5-pro1344.35364007949833.4454571461253840.00.01.002051759345898.45129250.0128102400.9276983017835836{'min_latency_ms': 615.3252124786377, 'max_latency_ms': 1981.612205505371, 'p95_latency_ms': np.float64(1803.935217857361), 'p99_latency_ms': np.float64(1946.0768079757688), 'total_time_s': 5.8047449588775635, 'initial_memory_mb': 294.4296875, 'final_memory_mb': 294.4296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00064, 'quality_std': 0.05905363250623063, 'data_size_processed': 1000, 'model_provider': 'gemini'}0.00.00.00.00False00.000.0False
2916scaling_testgemini-1.5-pro1202.21995115280153.6968698314009320.00.01.002051759345907.57072640.0128102400.9307740387961949{'min_latency_ms': 589.9953842163086, 'max_latency_ms': 1967.3075675964355, 'p95_latency_ms': np.float64(1913.6008977890015), 'p99_latency_ms': np.float64(1956.5662336349487), 'total_time_s': 5.409982204437256, 'initial_memory_mb': 294.4296875, 'final_memory_mb': 294.4296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00064, 'quality_std': 0.04978369465928124, 'data_size_processed': 1000, 'model_provider': 'gemini'}0.00.00.00.00False00.000.0False
301scaling_testgemini-1.5-flash1053.95122766494753.8232652803761660.00.01.002051759345915.09478190.007679999999999998102400.8813998853517441{'min_latency_ms': -36.76271438598633, 'max_latency_ms': 1967.0710563659668, 'p95_latency_ms': np.float64(1855.4362535476685), 'p99_latency_ms': np.float64(1944.744095802307), 'total_time_s': 5.231130599975586, 'initial_memory_mb': 294.4296875, 'final_memory_mb': 294.4296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0003839999999999999, 'quality_std': 0.050008698196664016, 'data_size_processed': 1000, 'model_provider': 'gemini'}0.00.00.00.00False00.000.0False
316scaling_testgemini-1.5-flash1155.39114475250243.6156368667199920.00.01.002051759345925.06945630.007679999999999998102400.9025102091839412{'min_latency_ms': 502.6116371154785, 'max_latency_ms': 1947.0453262329102, 'p95_latency_ms': np.float64(1765.414369106293), 'p99_latency_ms': np.float64(1910.7191348075864), 'total_time_s': 5.531528949737549, 'initial_memory_mb': 294.4296875, 'final_memory_mb': 294.4296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0003839999999999999, 'quality_std': 0.059194105459554974, 'data_size_processed': 1000, 'model_provider': 'gemini'}0.00.00.00.00False00.000.0False
3211scaling_testgemini-1.5-flash1217.66122579574583.7569650866731010.00.01.002051759345934.11833830.007679999999999998102400.8709830012564668{'min_latency_ms': 560.8868598937988, 'max_latency_ms': 2007.932424545288, 'p95_latency_ms': np.float64(1776.0017752647402), 'p99_latency_ms': np.float64(1961.5462946891782), 'total_time_s': 5.323445796966553, 'initial_memory_mb': 294.4296875, 'final_memory_mb': 294.4296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0003839999999999999, 'quality_std': 0.052873446152615404, 'data_size_processed': 1000, 'model_provider': 'gemini'}0.00.00.00.00False00.000.0False
3316scaling_testgemini-1.5-flash1351.52283906936653.3679959904962590.00.01.002051759345942.20997880.007679999999999998102400.872315613940513{'min_latency_ms': 689.1014575958252, 'max_latency_ms': 1980.147361755371, 'p95_latency_ms': np.float64(1956.2964797019958), 'p99_latency_ms': np.float64(1975.377185344696), 'total_time_s': 5.938249349594116, 'initial_memory_mb': 294.4296875, 'final_memory_mb': 294.4296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0003839999999999999, 'quality_std': 0.05361394744479093, 'data_size_processed': 1000, 'model_provider': 'gemini'}0.00.00.00.00False00.000.0False
341scaling_testllama-3.1-8b1306.5912365913393.30700392613205940.00.01.002051759345952.86929350.002048000000000001102400.7778348786353027{'min_latency_ms': 555.4070472717285, 'max_latency_ms': 1988.0244731903076, 'p95_latency_ms': np.float64(1957.3988199234009), 'p99_latency_ms': np.float64(1981.8993425369263), 'total_time_s': 6.047770261764526, 'initial_memory_mb': 294.4296875, 'final_memory_mb': 294.4296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00010240000000000006, 'quality_std': 0.05832225784189981, 'data_size_processed': 1000, 'model_provider': 'llama'}0.00.00.00.00False00.000.0False
356scaling_testllama-3.1-8b1199.62228536605833.6343580862202390.00.01.002051759345963.51526470.002048000000000001102400.7696592403957419{'min_latency_ms': 541.0621166229248, 'max_latency_ms': 1914.41011428833, 'p95_latency_ms': np.float64(1768.0468797683716), 'p99_latency_ms': np.float64(1885.1374673843382), 'total_time_s': 5.503035068511963, 'initial_memory_mb': 294.4296875, 'final_memory_mb': 294.4296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00010240000000000006, 'quality_std': 0.06176209698043544, 'data_size_processed': 1000, 'model_provider': 'llama'}0.00.00.00.00False00.000.0False
3611scaling_testllama-3.1-8b1143.3585524559024.1739162971507520.00.01.002051759345973.84061810.002048000000000001102400.7857043630038748{'min_latency_ms': 631.817102432251, 'max_latency_ms': 1720.1111316680908, 'p95_latency_ms': np.float64(1547.544610500336), 'p99_latency_ms': np.float64(1685.5978274345396), 'total_time_s': 4.791662931442261, 'initial_memory_mb': 294.4296875, 'final_memory_mb': 294.4296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00010240000000000006, 'quality_std': 0.06142254552174686, 'data_size_processed': 1000, 'model_provider': 'llama'}0.00.00.00.00False00.000.0False
3716scaling_testllama-3.1-8b1228.60485315322883.6134651351302690.00.01.002051759345982.27595450.002048000000000001102400.7706622409066766{'min_latency_ms': 539.0913486480713, 'max_latency_ms': 1971.7633724212646, 'p95_latency_ms': np.float64(1819.2362308502197), 'p99_latency_ms': np.float64(1941.2579441070554), 'total_time_s': 5.534853458404541, 'initial_memory_mb': 294.4296875, 'final_memory_mb': 294.4296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00010240000000000006, 'quality_std': 0.05320944570994387, 'data_size_processed': 1000, 'model_provider': 'llama'}0.00.00.00.00False00.000.0False
381scaling_testllama-3.1-70b1424.07245635986332.9893942639007630.00.01.002051759345993.49491260.008192000000000005102400.8731561293258354{'min_latency_ms': 700.6974220275879, 'max_latency_ms': 1959.3937397003174, 'p95_latency_ms': np.float64(1924.493396282196), 'p99_latency_ms': np.float64(1952.4136710166931), 'total_time_s': 6.690318584442139, 'initial_memory_mb': 294.4296875, 'final_memory_mb': 294.4296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00040960000000000025, 'quality_std': 0.0352234743129485, 'data_size_processed': 1000, 'model_provider': 'llama'}0.00.00.00.00False00.000.0False
396scaling_testllama-3.1-70b1090.0035142898564.1459172075663530.00.01.002051759346002.33539320.008192000000000005102400.8796527768140011{'min_latency_ms': 508.23211669921875, 'max_latency_ms': 1798.6392974853516, 'p95_latency_ms': np.float64(1785.5579257011414), 'p99_latency_ms': np.float64(1796.0230231285095), 'total_time_s': 4.824023008346558, 'initial_memory_mb': 294.4296875, 'final_memory_mb': 294.4296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00040960000000000025, 'quality_std': 0.06407982743031454, 'data_size_processed': 1000, 'model_provider': 'llama'}0.00.00.00.00False00.000.0False
4011scaling_testllama-3.1-70b964.36669826507574.703926450905850.00.01.002051759346010.69742160.008192000000000005102400.8992009479579495{'min_latency_ms': 135.56504249572754, 'max_latency_ms': 1794.3906784057617, 'p95_latency_ms': np.float64(1775.5030393600464), 'p99_latency_ms': np.float64(1790.6131505966187), 'total_time_s': 4.251767158508301, 'initial_memory_mb': 294.4296875, 'final_memory_mb': 294.4296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00040960000000000025, 'quality_std': 0.050182727925105516, 'data_size_processed': 1000, 'model_provider': 'llama'}0.00.00.00.00False00.000.0False
4116scaling_testllama-3.1-70b1258.94768238067633.6538316041105150.1250.01.002051759346020.3880940.008192000000000005102400.8930892849911802{'min_latency_ms': 620.0413703918457, 'max_latency_ms': 1916.384220123291, 'p95_latency_ms': np.float64(1765.2448296546936), 'p99_latency_ms': np.float64(1886.1563420295713), 'total_time_s': 5.473706007003784, 'initial_memory_mb': 294.4296875, 'final_memory_mb': 294.5546875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00040960000000000025, 'quality_std': 0.04969618373257882, 'data_size_processed': 1000, 'model_provider': 'llama'}0.00.00.00.00False00.000.0False
425concurrent_testgpt-4o-mini1273.7020969390870.78510867969266110.00.01.001011759346033.23738840.000768000000000000151200.8342026655690804{'min_latency_ms': 741.3482666015625, 'max_latency_ms': 1817.1906471252441, 'p95_latency_ms': np.float64(1794.5520520210266), 'p99_latency_ms': np.float64(1812.6629281044006), 'total_time_s': 12.737090110778809, 'initial_memory_mb': 294.5546875, 'final_memory_mb': 294.5546875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 7.680000000000001e-05, 'quality_std': 0.0446055902590032, 'data_size_processed': 1000, 'model_provider': 'gpt'}0.00.00.00.00False00.000.0False
435concurrent_testgpt-4o-mini1511.3994836807252.9337631024401560.250.01.001061759346036.6472140.000768000000000000151200.8471277213854321{'min_latency_ms': 800.0023365020752, 'max_latency_ms': 1982.2335243225098, 'p95_latency_ms': np.float64(1942.5656914710999), 'p99_latency_ms': np.float64(1974.2999577522278), 'total_time_s': 3.4085915088653564, 'initial_memory_mb': 294.5546875, 'final_memory_mb': 294.8046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 7.680000000000001e-05, 'quality_std': 0.06432848764341552, 'data_size_processed': 1000, 'model_provider': 'gpt'}0.00.00.00.00False00.000.0False
445concurrent_testgpt-4o1150.04916191101070.86952289001328530.00.01.001011759346048.25873330.025651200.9599583095352598{'min_latency_ms': 544.191837310791, 'max_latency_ms': 1584.9177837371826, 'p95_latency_ms': np.float64(1511.2051010131834), 'p99_latency_ms': np.float64(1570.1752471923828), 'total_time_s': 11.50055980682373, 'initial_memory_mb': 294.8046875, 'final_memory_mb': 294.8046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00256, 'quality_std': 0.057087428808928614, 'data_size_processed': 1000, 'model_provider': 'gpt'}0.00.00.00.00False00.000.0False
455concurrent_testgpt-4o1241.90819263458253.229810297435190.00.01.001061759346051.35637570.025651200.9585199558650109{'min_latency_ms': 644.8915004730225, 'max_latency_ms': 1933.1202507019043, 'p95_latency_ms': np.float64(1865.2720570564268), 'p99_latency_ms': np.float64(1919.5506119728088), 'total_time_s': 3.0961570739746094, 'initial_memory_mb': 294.8046875, 'final_memory_mb': 294.8046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00256, 'quality_std': 0.04062204558012218, 'data_size_processed': 1000, 'model_provider': 'gpt'}0.00.00.00.00False00.000.0False
465concurrent_testgpt-4-turbo1581.87503814697270.63215811790296060.00.01.001011759346067.30179640.051251200.9324427514695872{'min_latency_ms': 833.935022354126, 'max_latency_ms': 2019.5622444152832, 'p95_latency_ms': np.float64(1978.4671545028687), 'p99_latency_ms': np.float64(2011.3432264328003), 'total_time_s': 15.818827152252197, 'initial_memory_mb': 294.8046875, 'final_memory_mb': 294.8046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00512, 'quality_std': 0.04654046504268862, 'data_size_processed': 1000, 'model_provider': 'gpt'}0.00.00.00.00False00.000.0False
475concurrent_testgpt-4-turbo1153.4322977066043.21689932402458470.00.01.001061759346070.41167620.051251200.9790878168553954{'min_latency_ms': 635.2591514587402, 'max_latency_ms': 1833.7628841400146, 'p95_latency_ms': np.float64(1808.298635482788), 'p99_latency_ms': np.float64(1828.6700344085693), 'total_time_s': 3.108583450317383, 'initial_memory_mb': 294.8046875, 'final_memory_mb': 294.8046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00512, 'quality_std': 0.038783270511690816, 'data_size_processed': 1000, 'model_provider': 'gpt'}0.00.00.00.00False00.000.0False
485concurrent_testclaude-3-5-sonnet1397.67837524414060.71546801027074220.00.01.001011759346084.50178240.01535999999999999951200.9421283071854264{'min_latency_ms': 532.8092575073242, 'max_latency_ms': 2028.5301208496094, 'p95_latency_ms': np.float64(1968.815779685974), 'p99_latency_ms': np.float64(2016.5872526168823), 'total_time_s': 13.976865291595459, 'initial_memory_mb': 294.8046875, 'final_memory_mb': 294.8046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0015359999999999998, 'quality_std': 0.041911119259679885, 'data_size_processed': 1000, 'model_provider': 'claude'}0.00.00.00.00False00.000.0False
495concurrent_testclaude-3-5-sonnet1215.261983871463.62784219839952330.00.01.001061759346087.25962160.01535999999999999951200.9131170426955485{'min_latency_ms': 568.2053565979004, 'max_latency_ms': 1612.9648685455322, 'p95_latency_ms': np.float64(1559.6276402473447), 'p99_latency_ms': np.float64(1602.2974228858948), 'total_time_s': 2.7564594745635986, 'initial_memory_mb': 294.8046875, 'final_memory_mb': 294.8046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0015359999999999998, 'quality_std': 0.04319876804321411, 'data_size_processed': 1000, 'model_provider': 'claude'}0.00.00.00.00False00.000.0False
505concurrent_testclaude-3-haiku1299.22769069671630.76968261903313950.00.01.001011759346100.3644070.012851200.8252745814485088{'min_latency_ms': 668.3671474456787, 'max_latency_ms': 2041.351318359375, 'p95_latency_ms': np.float64(1843.0875778198238), 'p99_latency_ms': np.float64(2001.6985702514648), 'total_time_s': 12.992368221282959, 'initial_memory_mb': 294.8046875, 'final_memory_mb': 294.8046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00128, 'quality_std': 0.058205855327116265, 'data_size_processed': 1000, 'model_provider': 'claude'}0.00.00.00.00False00.000.0False
515concurrent_testclaude-3-haiku1297.5081920623783.65816546443210870.00.01.001061759346103.09939960.012851200.8496515913760503{'min_latency_ms': 649.4293212890625, 'max_latency_ms': 1873.1675148010254, 'p95_latency_ms': np.float64(1843.8988208770752), 'p99_latency_ms': np.float64(1867.3137760162354), 'total_time_s': 2.7336106300354004, 'initial_memory_mb': 294.8046875, 'final_memory_mb': 294.8046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00128, 'quality_std': 0.06872259975771335, 'data_size_processed': 1000, 'model_provider': 'claude'}0.00.00.00.00False00.000.0False
525concurrent_testclaude-3-sonnet1239.81237411499020.80656922052638740.00.01.001011759346114.96500350.0768000000000000151200.8917269647002374{'min_latency_ms': 559.9334239959717, 'max_latency_ms': 1828.9196491241455, 'p95_latency_ms': np.float64(1804.089903831482), 'p99_latency_ms': np.float64(1823.9537000656128), 'total_time_s': 12.398191928863525, 'initial_memory_mb': 294.8046875, 'final_memory_mb': 294.8046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.007680000000000001, 'quality_std': 0.06728256480558785, 'data_size_processed': 1000, 'model_provider': 'claude'}0.00.00.00.00False00.000.0False
535concurrent_testclaude-3-sonnet1325.38752555847173.23056132904009450.00.01.001061759346118.0621730.0768000000000000151200.8904253939966993{'min_latency_ms': 598.4294414520264, 'max_latency_ms': 1956.3815593719482, 'p95_latency_ms': np.float64(1906.8223834037778), 'p99_latency_ms': np.float64(1946.4697241783142), 'total_time_s': 3.0954372882843018, 'initial_memory_mb': 294.8046875, 'final_memory_mb': 294.8046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.007680000000000001, 'quality_std': 0.06220445402424322, 'data_size_processed': 1000, 'model_provider': 'claude'}0.00.00.00.00False00.000.0False
545concurrent_testgemini-1.5-pro1264.27545547485350.79096302178324750.00.01.001011759346130.82829640.006451200.8998460053229075{'min_latency_ms': 532.9890251159668, 'max_latency_ms': 1795.492172241211, 'p95_latency_ms': np.float64(1745.6329107284544), 'p99_latency_ms': np.float64(1785.5203199386597), 'total_time_s': 12.642816066741943, 'initial_memory_mb': 294.8046875, 'final_memory_mb': 294.8046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00064, 'quality_std': 0.04050886994282564, 'data_size_processed': 1000, 'model_provider': 'gemini'}0.00.00.00.00False00.000.0False
555concurrent_testgemini-1.5-pro1342.90063381195073.78291501811230150.00.01.001061759346133.4729560.006451200.9029938738274873{'min_latency_ms': 701.9498348236084, 'max_latency_ms': 1964.576005935669, 'p95_latency_ms': np.float64(1872.5560665130613), 'p99_latency_ms': np.float64(1946.1720180511475), 'total_time_s': 2.6434640884399414, 'initial_memory_mb': 294.8046875, 'final_memory_mb': 294.8046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00064, 'quality_std': 0.05723923041822323, 'data_size_processed': 1000, 'model_provider': 'gemini'}0.00.00.00.00False00.000.0False
565concurrent_testgemini-1.5-flash1368.25885772705080.73085159070935060.00.01.001011759346147.27175740.003839999999999999751200.8795901650694117{'min_latency_ms': 620.3913688659668, 'max_latency_ms': 2018.2685852050781, 'p95_latency_ms': np.float64(1993.7742233276367), 'p99_latency_ms': np.float64(2013.3697128295898), 'total_time_s': 13.682668447494507, 'initial_memory_mb': 294.8046875, 'final_memory_mb': 294.8046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00038399999999999996, 'quality_std': 0.05927449072307118, 'data_size_processed': 1000, 'model_provider': 'gemini'}0.00.00.00.00False00.000.0False
575concurrent_testgemini-1.5-flash1207.86297321319583.28795928243020440.00.01.001061759346150.3146170.003839999999999999751200.8611774574826484{'min_latency_ms': 594.973087310791, 'max_latency_ms': 1811.2657070159912, 'p95_latency_ms': np.float64(1681.6352963447569), 'p99_latency_ms': np.float64(1785.3396248817444), 'total_time_s': 3.041400194168091, 'initial_memory_mb': 294.8046875, 'final_memory_mb': 294.8046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00038399999999999996, 'quality_std': 0.07904328865026665, 'data_size_processed': 1000, 'model_provider': 'gemini'}0.00.00.00.00False00.000.0False
585concurrent_testllama-3.1-8b1144.29101943969730.87389036312763320.00.01.001011759346161.8823890.001024000000000000251200.7805684315735588{'min_latency_ms': 594.846248626709, 'max_latency_ms': 1759.0994834899902, 'p95_latency_ms': np.float64(1631.7564606666563), 'p99_latency_ms': np.float64(1733.6308789253235), 'total_time_s': 11.443083047866821, 'initial_memory_mb': 294.8046875, 'final_memory_mb': 294.8046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00010240000000000002, 'quality_std': 0.0613021253594286, 'data_size_processed': 1000, 'model_provider': 'llama'}0.00.00.00.00False00.000.0False
595concurrent_testllama-3.1-8b1128.6666154861453.5270063839738530.00.01.001061759346164.71909070.001024000000000000251200.7915276538063776{'min_latency_ms': 610.3026866912842, 'max_latency_ms': 1934.2899322509766, 'p95_latency_ms': np.float64(1909.2738270759583), 'p99_latency_ms': np.float64(1929.286711215973), 'total_time_s': 2.835265636444092, 'initial_memory_mb': 294.8046875, 'final_memory_mb': 294.8046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00010240000000000002, 'quality_std': 0.055242108041169316, 'data_size_processed': 1000, 'model_provider': 'llama'}0.00.00.00.00False00.000.0False
605concurrent_testllama-3.1-70b1341.4107322692870.74548053633454770.00.01.001011759346178.25718240.00409600000000000151200.8513858389112968{'min_latency_ms': 566.3845539093018, 'max_latency_ms': 1769.1750526428223, 'p95_latency_ms': np.float64(1743.9924359321594), 'p99_latency_ms': np.float64(1764.1385293006897), 'total_time_s': 13.414166450500488, 'initial_memory_mb': 294.8046875, 'final_memory_mb': 294.8046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0004096000000000001, 'quality_std': 0.06286695897481548, 'data_size_processed': 1000, 'model_provider': 'llama'}0.00.00.00.00False00.000.0False
615concurrent_testllama-3.1-70b1410.38112640380863.520227883404470.00.01.001061759346181.09923080.00409600000000000151200.8534058400920448{'min_latency_ms': 572.9773044586182, 'max_latency_ms': 1928.0850887298584, 'p95_latency_ms': np.float64(1903.529143333435), 'p99_latency_ms': np.float64(1923.1738996505737), 'total_time_s': 2.8407251834869385, 'initial_memory_mb': 294.8046875, 'final_memory_mb': 294.8046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0004096000000000001, 'quality_std': 0.059750620144052545, 'data_size_processed': 1000, 'model_provider': 'llama'}0.00.00.00.00False00.000.0False
625memory_testgpt-4o-mini1177.24404811859133.975010087017980.00.01.005051759346193.79012010.0038400000000000023256000.8512259391579574{'min_latency_ms': 537.5485420227051, 'max_latency_ms': 2001.0862350463867, 'p95_latency_ms': np.float64(1892.5400853157041), 'p99_latency_ms': np.float64(1985.4257130622864), 'total_time_s': 12.578584432601929, 'initial_memory_mb': 294.8046875, 'final_memory_mb': 294.8046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 7.680000000000005e-05, 'quality_std': 0.0581968026848211, 'data_size_processed': 1000, 'model_provider': 'gpt', 'iteration': 0}0.00.00.00.00False00.000.0False
635memory_testgpt-4o-mini1229.80267524719243.92823696794603630.00.01.005051759346206.63009050.0038400000000000023256000.8537868196468017{'min_latency_ms': 518.6026096343994, 'max_latency_ms': 1944.331407546997, 'p95_latency_ms': np.float64(1909.6850633621214), 'p99_latency_ms': np.float64(1940.652117729187), 'total_time_s': 12.72835636138916, 'initial_memory_mb': 294.8046875, 'final_memory_mb': 294.8046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 7.680000000000005e-05, 'quality_std': 0.05181407518487485, 'data_size_processed': 1000, 'model_provider': 'gpt', 'iteration': 1}0.00.00.00.00False00.000.0False
645memory_testgpt-4o-mini1274.81444835662843.74831199667098240.00.01.005051759346220.09000730.0038400000000000023256000.8487480924622282{'min_latency_ms': 529.292106628418, 'max_latency_ms': 1996.4158535003662, 'p95_latency_ms': np.float64(1960.6919050216675), 'p99_latency_ms': np.float64(1988.2149648666382), 'total_time_s': 13.339337825775146, 'initial_memory_mb': 294.8046875, 'final_memory_mb': 294.8046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 7.680000000000005e-05, 'quality_std': 0.05812899461310237, 'data_size_processed': 1000, 'model_provider': 'gpt', 'iteration': 2}0.00.00.00.00False00.000.0False
655memory_testgpt-4o1174.50570106506354.05141363899861150.00.01.005051759346232.5577840.12800000000000017256000.9484191580718665{'min_latency_ms': 286.58127784729004, 'max_latency_ms': 1877.345085144043, 'p95_latency_ms': np.float64(1735.1435780525208), 'p99_latency_ms': np.float64(1842.000467777252), 'total_time_s': 12.341371297836304, 'initial_memory_mb': 294.8046875, 'final_memory_mb': 294.8046875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0025600000000000032, 'quality_std': 0.0491398572941036, 'data_size_processed': 1000, 'model_provider': 'gpt', 'iteration': 0}0.00.00.00.00False00.000.0False
665memory_testgpt-4o1225.3885936737063.8759324296331760.1250.01.005051759346245.56695340.12800000000000017256000.9557179217710832{'min_latency_ms': 514.6803855895996, 'max_latency_ms': 2034.6620082855225, 'p95_latency_ms': np.float64(1909.4360709190366), 'p99_latency_ms': np.float64(2010.34743309021), 'total_time_s': 12.900121688842773, 'initial_memory_mb': 294.8046875, 'final_memory_mb': 294.9296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0025600000000000032, 'quality_std': 0.04870463047338363, 'data_size_processed': 1000, 'model_provider': 'gpt', 'iteration': 1}0.00.00.00.00False00.000.0False
675memory_testgpt-4o1244.00219917297363.72664461015467770.00.01.005051759346259.14147760.12800000000000017256000.9458944372937584{'min_latency_ms': 521.9912528991699, 'max_latency_ms': 1986.6855144500732, 'p95_latency_ms': np.float64(1953.3554077148438), 'p99_latency_ms': np.float64(1978.9683985710144), 'total_time_s': 13.416895151138306, 'initial_memory_mb': 294.9296875, 'final_memory_mb': 294.9296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0025600000000000032, 'quality_std': 0.04851286804634898, 'data_size_processed': 1000, 'model_provider': 'gpt', 'iteration': 2}0.00.00.00.00False00.000.0False
685memory_testgpt-4-turbo1181.36153221130374.1249984166032190.00.01.005051759346271.3745780.25600000000000034256000.9651345363111258{'min_latency_ms': 353.2071113586426, 'max_latency_ms': 1966.524362564087, 'p95_latency_ms': np.float64(1945.0057744979858), 'p99_latency_ms': np.float64(1965.7717752456665), 'total_time_s': 12.121216773986816, 'initial_memory_mb': 294.9296875, 'final_memory_mb': 294.9296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0051200000000000065, 'quality_std': 0.04338778763022959, 'data_size_processed': 1000, 'model_provider': 'gpt', 'iteration': 0}0.00.00.00.00False00.000.0False
695memory_testgpt-4-turbo1291.40556812286383.775524009521120.00.01.005051759346284.7318120.25600000000000034256000.9689389907566063{'min_latency_ms': 555.095911026001, 'max_latency_ms': 2027.0910263061523, 'p95_latency_ms': np.float64(1966.5393114089964), 'p99_latency_ms': np.float64(2018.9284563064575), 'total_time_s': 13.243194818496704, 'initial_memory_mb': 294.9296875, 'final_memory_mb': 294.9296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0051200000000000065, 'quality_std': 0.04154143035607859, 'data_size_processed': 1000, 'model_provider': 'gpt', 'iteration': 1}0.00.00.00.00False00.000.0False
705memory_testgpt-4-turbo1261.42082691192633.6632083211300740.00.01.005051759346298.49054930.25600000000000034256000.9573488473081913{'min_latency_ms': 284.8320007324219, 'max_latency_ms': 2011.866807937622, 'p95_latency_ms': np.float64(1975.5298137664795), 'p99_latency_ms': np.float64(2000.7115292549133), 'total_time_s': 13.649237394332886, 'initial_memory_mb': 294.9296875, 'final_memory_mb': 294.9296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0051200000000000065, 'quality_std': 0.04380501534660363, 'data_size_processed': 1000, 'model_provider': 'gpt', 'iteration': 2}0.00.00.00.00False00.000.0False
715memory_testclaude-3-5-sonnet1270.35431385040283.79443209890906140.00.01.005051759346311.79360220.07680000000000001256000.948463600922609{'min_latency_ms': 622.9770183563232, 'max_latency_ms': 1970.0510501861572, 'p95_latency_ms': np.float64(1868.455410003662), 'p99_latency_ms': np.float64(1957.5506472587585), 'total_time_s': 13.177202463150024, 'initial_memory_mb': 294.9296875, 'final_memory_mb': 294.9296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.001536, 'quality_std': 0.04872900892927657, 'data_size_processed': 1000, 'model_provider': 'claude', 'iteration': 0}0.00.00.00.00False00.000.0False
725memory_testclaude-3-5-sonnet1154.5276212692264.1078021488183130.00.01.005051759346324.07820340.07680000000000001256000.9535056752128789{'min_latency_ms': 526.8404483795166, 'max_latency_ms': 1841.3877487182617, 'p95_latency_ms': np.float64(1815.3946280479431), 'p99_latency_ms': np.float64(1837.1384692192078), 'total_time_s': 12.171959161758423, 'initial_memory_mb': 294.9296875, 'final_memory_mb': 294.9296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.001536, 'quality_std': 0.04600056992617095, 'data_size_processed': 1000, 'model_provider': 'claude', 'iteration': 1}0.00.00.00.00False00.000.0False
735memory_testclaude-3-5-sonnet1341.66581630706793.50503254939778050.00.01.005051759346338.45605730.07680000000000001256000.947231761746643{'min_latency_ms': 607.1841716766357, 'max_latency_ms': 1968.3496952056885, 'p95_latency_ms': np.float64(1938.420307636261), 'p99_latency_ms': np.float64(1963.8122081756592), 'total_time_s': 14.265202760696411, 'initial_memory_mb': 294.9296875, 'final_memory_mb': 294.9296875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.001536, 'quality_std': 0.0468041040494112, 'data_size_processed': 1000, 'model_provider': 'claude', 'iteration': 2}0.00.00.00.00False00.000.0False
745memory_testclaude-3-haiku1268.90418052673343.65274057349026070.1250.01.005051759346352.27602840.06400000000000008256000.8657832919908838{'min_latency_ms': 576.9007205963135, 'max_latency_ms': 1978.3263206481934, 'p95_latency_ms': np.float64(1900.9657382965088), 'p99_latency_ms': np.float64(1977.4397349357605), 'total_time_s': 13.688352346420288, 'initial_memory_mb': 294.9296875, 'final_memory_mb': 295.0546875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0012800000000000016, 'quality_std': 0.05791027367020173, 'data_size_processed': 1000, 'model_provider': 'claude', 'iteration': 0}0.00.00.00.00False00.000.0False
755memory_testclaude-3-haiku1273.69898319244383.76025437774308770.00.01.005051759346365.6818290.06400000000000008256000.8396294693060197{'min_latency_ms': 521.7316150665283, 'max_latency_ms': 1988.7199401855469, 'p95_latency_ms': np.float64(1945.9344744682312), 'p99_latency_ms': np.float64(1987.1683859825134), 'total_time_s': 13.296972751617432, 'initial_memory_mb': 295.0546875, 'final_memory_mb': 295.0546875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0012800000000000016, 'quality_std': 0.06291349263235946, 'data_size_processed': 1000, 'model_provider': 'claude', 'iteration': 1}0.00.00.00.00False00.000.0False
765memory_testclaude-3-haiku1234.92699146270753.93350823453181240.00.01.005051759346378.51926640.06400000000000008256000.8469784358915146{'min_latency_ms': 529.503345489502, 'max_latency_ms': 1981.7008972167969, 'p95_latency_ms': np.float64(1859.1547846794128), 'p99_latency_ms': np.float64(1963.3227896690369), 'total_time_s': 12.711299180984497, 'initial_memory_mb': 295.0546875, 'final_memory_mb': 295.0546875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0012800000000000016, 'quality_std': 0.061722943046806616, 'data_size_processed': 1000, 'model_provider': 'claude', 'iteration': 2}0.00.00.00.00False00.000.0False
775memory_testclaude-3-sonnet1195.90081691741944.069627383824440.00.01.005051759346390.91448970.3840000000000003256000.9026531444228556{'min_latency_ms': -36.6673469543457, 'max_latency_ms': 1991.610050201416, 'p95_latency_ms': np.float64(1819.4202184677124), 'p99_latency_ms': np.float64(1987.222683429718), 'total_time_s': 12.286137104034424, 'initial_memory_mb': 295.0546875, 'final_memory_mb': 295.0546875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.007680000000000005, 'quality_std': 0.058229589360407986, 'data_size_processed': 1000, 'model_provider': 'claude', 'iteration': 0}0.00.00.00.00False00.000.0False
785memory_testclaude-3-sonnet1372.03798294067383.5022533454658050.00.01.005051759346405.30434940.3840000000000003256000.8837364473272626{'min_latency_ms': 543.1270599365234, 'max_latency_ms': 1992.779016494751, 'p95_latency_ms': np.float64(1931.822681427002), 'p99_latency_ms': np.float64(1987.4089169502258), 'total_time_s': 14.276522874832153, 'initial_memory_mb': 295.0546875, 'final_memory_mb': 295.0546875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.007680000000000005, 'quality_std': 0.05634614113838598, 'data_size_processed': 1000, 'model_provider': 'claude', 'iteration': 1}0.00.00.00.00False00.000.0False
795memory_testclaude-3-sonnet1257.27090358734133.77648570621827060.00.01.005051759346418.65218540.3840000000000003256000.9053414058751514{'min_latency_ms': 529.8404693603516, 'max_latency_ms': 1990.1280403137207, 'p95_latency_ms': np.float64(1911.1806631088257), 'p99_latency_ms': np.float64(1976.6331052780151), 'total_time_s': 13.239822387695312, 'initial_memory_mb': 295.0546875, 'final_memory_mb': 295.0546875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.007680000000000005, 'quality_std': 0.050506656009957705, 'data_size_processed': 1000, 'model_provider': 'claude', 'iteration': 2}0.00.00.00.00False00.000.0False
805memory_testgemini-1.5-pro1221.59514904022223.83729089698453230.00.01.005051759346431.79215650.03200000000000004256000.9365925291921394{'min_latency_ms': 329.1811943054199, 'max_latency_ms': 1995.384693145752, 'p95_latency_ms': np.float64(1965.0332808494568), 'p99_latency_ms': np.float64(1988.3063769340515), 'total_time_s': 13.030025959014893, 'initial_memory_mb': 295.0546875, 'final_memory_mb': 295.0546875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0006400000000000008, 'quality_std': 0.04847128641002876, 'data_size_processed': 1000, 'model_provider': 'gemini', 'iteration': 0}0.00.00.00.00False00.000.0False
815memory_testgemini-1.5-pro1351.83554649353033.62279754365526060.00.01.005051759346445.71264480.03200000000000004256000.9323552590826123{'min_latency_ms': 515.129566192627, 'max_latency_ms': 2008.0702304840088, 'p95_latency_ms': np.float64(1958.6564779281616), 'p99_latency_ms': np.float64(2004.1296029090881), 'total_time_s': 13.801488876342773, 'initial_memory_mb': 295.0546875, 'final_memory_mb': 295.0546875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0006400000000000008, 'quality_std': 0.055840796126395656, 'data_size_processed': 1000, 'model_provider': 'gemini', 'iteration': 1}0.00.00.00.00False00.000.0False
825memory_testgemini-1.5-pro1240.6225347518923.88133840983744530.00.01.005051759346458.71927290.03200000000000004256000.9407390543744837{'min_latency_ms': -29.146671295166016, 'max_latency_ms': 1934.4398975372314, 'p95_latency_ms': np.float64(1849.7230291366577), 'p99_latency_ms': np.float64(1918.0084466934204), 'total_time_s': 12.8821542263031, 'initial_memory_mb': 295.0546875, 'final_memory_mb': 295.0546875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0006400000000000008, 'quality_std': 0.050597003908357786, 'data_size_processed': 1000, 'model_provider': 'gemini', 'iteration': 2}0.00.00.00.00False00.000.0False
835memory_testgemini-1.5-flash1237.67026424407963.8129234956443460.00.01.005051759346471.95889740.019200000000000002256000.8556073429019542{'min_latency_ms': 536.4787578582764, 'max_latency_ms': 2010.1728439331055, 'p95_latency_ms': np.float64(1911.8669629096985), 'p99_latency_ms': np.float64(1976.080708503723), 'total_time_s': 13.113297462463379, 'initial_memory_mb': 295.0546875, 'final_memory_mb': 295.0546875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.000384, 'quality_std': 0.06082135675952047, 'data_size_processed': 1000, 'model_provider': 'gemini', 'iteration': 0}0.00.00.00.00False00.000.0False
845memory_testgemini-1.5-flash1180.09808063507084.0160490908320030.00.01.005051759346484.53277440.019200000000000002256000.8718428063415768{'min_latency_ms': 109.58051681518555, 'max_latency_ms': 1993.358850479126, 'p95_latency_ms': np.float64(1872.3165988922117), 'p99_latency_ms': np.float64(1992.416422367096), 'total_time_s': 12.450047016143799, 'initial_memory_mb': 295.0546875, 'final_memory_mb': 295.0546875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.000384, 'quality_std': 0.0613916834940056, 'data_size_processed': 1000, 'model_provider': 'gemini', 'iteration': 1}0.00.00.00.00False00.000.0False
855memory_testgemini-1.5-flash1194.44900989532474.0099361194830760.00.01.005051759346497.12010880.019200000000000002256000.8652112059805899{'min_latency_ms': 520.3211307525635, 'max_latency_ms': 1942.4259662628174, 'p95_latency_ms': np.float64(1834.6370577812195), 'p99_latency_ms': np.float64(1890.3984904289243), 'total_time_s': 12.469026565551758, 'initial_memory_mb': 295.0546875, 'final_memory_mb': 295.0546875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.000384, 'quality_std': 0.05312368368226588, 'data_size_processed': 1000, 'model_provider': 'gemini', 'iteration': 2}0.00.00.00.00False00.000.0False
865memory_testllama-3.1-8b1306.20167732238773.6837635476965550.00.01.005051759346510.8127320.005119999999999998256000.7727309350554936{'min_latency_ms': 527.4953842163086, 'max_latency_ms': 1997.086524963379, 'p95_latency_ms': np.float64(1942.7793741226194), 'p99_latency_ms': np.float64(1994.0643763542175), 'total_time_s': 13.573075294494629, 'initial_memory_mb': 295.0546875, 'final_memory_mb': 295.0546875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00010239999999999995, 'quality_std': 0.05596283861854901, 'data_size_processed': 1000, 'model_provider': 'llama', 'iteration': 0}0.00.00.00.00False00.000.0False
875memory_testllama-3.1-8b1304.12514686584473.6173837447730050.00.01.005051759346524.77119370.005119999999999998256000.785787220179362{'min_latency_ms': 112.00571060180664, 'max_latency_ms': 2015.146255493164, 'p95_latency_ms': np.float64(2001.4938592910767), 'p99_latency_ms': np.float64(2012.321424484253), 'total_time_s': 13.822144269943237, 'initial_memory_mb': 295.0546875, 'final_memory_mb': 295.0546875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00010239999999999995, 'quality_std': 0.0552285639827787, 'data_size_processed': 1000, 'model_provider': 'llama', 'iteration': 1}0.00.00.00.00False00.000.0False
885memory_testllama-3.1-8b1290.53462982177733.6715227103110510.00.01.005051759346538.50841070.005119999999999998256000.7771978709125356{'min_latency_ms': 565.7510757446289, 'max_latency_ms': 1945.1093673706055, 'p95_latency_ms': np.float64(1906.785237789154), 'p99_latency_ms': np.float64(1942.4526476860046), 'total_time_s': 13.618327856063843, 'initial_memory_mb': 295.0546875, 'final_memory_mb': 295.0546875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.00010239999999999995, 'quality_std': 0.057252814774054535, 'data_size_processed': 1000, 'model_provider': 'llama', 'iteration': 2}0.00.00.00.00False00.000.0False
895memory_testllama-3.1-70b1213.93347263336183.9476752767374860.00.01.005051759346551.29517440.02047999999999999256000.8683286341213061{'min_latency_ms': -79.86569404602051, 'max_latency_ms': 2014.9149894714355, 'p95_latency_ms': np.float64(1919.9433565139768), 'p99_latency_ms': np.float64(1992.4925136566162), 'total_time_s': 12.665682077407837, 'initial_memory_mb': 295.0546875, 'final_memory_mb': 295.0546875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0004095999999999998, 'quality_std': 0.05862810413022958, 'data_size_processed': 1000, 'model_provider': 'llama', 'iteration': 0}0.00.00.00.00False00.000.0False
905memory_testllama-3.1-70b1298.19587707519533.70497118979767630.00.01.005051759346564.92800330.02047999999999999256000.8889975698232048{'min_latency_ms': 503.5574436187744, 'max_latency_ms': 2020.4124450683594, 'p95_latency_ms': np.float64(1901.4497756958008), 'p99_latency_ms': np.float64(1986.3133001327512), 'total_time_s': 13.495381593704224, 'initial_memory_mb': 295.0546875, 'final_memory_mb': 295.0546875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0004095999999999998, 'quality_std': 0.053463278827038344, 'data_size_processed': 1000, 'model_provider': 'llama', 'iteration': 1}0.00.00.00.00False00.000.0False
915memory_testllama-3.1-70b1187.0401382446294.1651391128126110.00.01.005051759346577.04679780.02047999999999999256000.8884529182459214{'min_latency_ms': 506.2377452850342, 'max_latency_ms': 2026.6106128692627, 'p95_latency_ms': np.float64(1958.3556652069092), 'p99_latency_ms': np.float64(2007.5032830238342), 'total_time_s': 12.004400968551636, 'initial_memory_mb': 295.0546875, 'final_memory_mb': 295.0546875, 'avg_tokens_per_request': 512.0, 'cost_per_request': 0.0004095999999999998, 'quality_std': 0.05625669416735748, 'data_size_processed': 1000, 'model_provider': 'llama', 'iteration': 2}0.00.00.00.00False00.000.0False