Spaces:
Running
Running
Task,1000,2000,4000,8000,16000,32000,64000,128000,avg,avg.L | |
glm-4-9b-chat,31.634064359831303,34.989931742075846,46.36923333133574,57.26961338964374,63.61295034310319,68.33923257857671,72.15825322852012,72.92730988128146,55.91257360679601,71.14159856279277 | |
Mistral-Nemo-Instruct,33.442943349483386,35.45098716112904,48.17141258583688,57.95087475137824,65.37779827704927,65.49015807503481,63.60618294804688,61.730166764004046,53.90256548899532,63.60883592902858 | |
Mistral-Large-Instruct-AWQ,49.14942052867251,51.23221381539887,60.7793522691761,71.94684170889438,77.09652619929832,79.45080285304805,77.76925084338332,62.18320059627511,66.20095110176834,73.13441809756883 | |
Llama-3.1-8B-Instruct,32.40198895196592,32.979539690235654,45.0106861254419,56.07951990889208,64.18069403777689,69.62343477623669,71.53647862646864,69.22900618587121,55.130168537861124,70.12963986285885 | |
Llama-3.1-70B-Instruct-AWQ,38.7480657800462,42.87159365068116,53.97630601173803,66.0747722354859,73.11797388808664,76.55703704352354,78.2577157921391,65.5267461115416,61.89127631415526,73.44716631573475 | |
Qwen2-7B-Instruct,31.530664252548508,35.5910761564275,44.63338088737434,54.35904179697225,63.05187841718273,64.46105522821793,66.9059768609517,63.06871149116235,52.95022313635466,64.81191452677733 | |
Qwen2-72B-Instruct-AWQ,36.40772920298174,41.89206817510168,54.241398436281806,65.33266907435282,73.39122500043426,76.52590896239253,77.50975041531201,77.46999427458152,62.8463429426798,77.16855121742869 | |
Phi-3-Mini-Instruct,30.26814474962816,30.90238348500332,38.09444166155989,48.140298057357256,53.58382886518213,57.28513647576864,56.83349679389524,48.723036083017135,45.47884577142647,54.28055645089368 | |
Phi-3-Medium-Instruct,31.72976117411381,33.54934540740912,39.10262687186865,49.82711345124473,58.29010994509822,61.17171452761886,60.62625650790335,45.31693119039325,47.45173238445625,55.70496740863848 | |
Phi-3-Small-Instruct,31.48315960873518,36.27405424169346,46.19687960118542,54.34055832812775,59.62975835311172,59.733406453060674,60.19785010451274,48.96685678550229,49.60281543449115,56.299371114358564 | |
Jamba-1.5-Mini,32.100142207922914,36.90890463687818,48.61257141384665,60.28550585861812,66.0507948010125,68.33428275575879,66.01898750117914,65.27696038728068,55.448518695312124,66.54341021473954 | |
Gemini-1.5-Pro,36.40,47.31,58.01,65.49,71.43,74.22,72.43,72.42,62.21,73.03 |