| { | |
| "prefill": { | |
| "memory": { | |
| "unit": "MB", | |
| "max_ram": 3602.481152, | |
| "max_vram": 13600.03072, | |
| "max_reserved": 13107.2, | |
| "max_allocated": 10874.392064 | |
| }, | |
| "latency": { | |
| "unit": "s", | |
| "mean": 0.3773231913248697, | |
| "stdev": 0.004032105846019261, | |
| "values": [ | |
| 0.39744345092773437, | |
| 0.38059417724609373, | |
| 0.3768924255371094, | |
| 0.37622784423828126, | |
| 0.37658932495117187, | |
| 0.37643670654296874, | |
| 0.37635992431640625, | |
| 0.3764459533691406, | |
| 0.37676953125, | |
| 0.3766363830566406, | |
| 0.3758919677734375, | |
| 0.3761131591796875, | |
| 0.3763240966796875, | |
| 0.3764326477050781, | |
| 0.37676953125, | |
| 0.3764981689453125, | |
| 0.37646951293945313, | |
| 0.3763261413574219, | |
| 0.3764234313964844, | |
| 0.37589297485351564, | |
| 0.3767470092773437, | |
| 0.3763292236328125, | |
| 0.3761827697753906, | |
| 0.3762247619628906, | |
| 0.37629031372070315, | |
| 0.37621145629882813, | |
| 0.3762032775878906 | |
| ] | |
| }, | |
| "throughput": { | |
| "unit": "tokens/s", | |
| "value": 848.0793318756936 | |
| }, | |
| "energy": null, | |
| "efficiency": null | |
| }, | |
| "decode": { | |
| "memory": { | |
| "unit": "MB", | |
| "max_ram": 3622.457344, | |
| "max_vram": 13816.037376, | |
| "max_reserved": 13323.206656, | |
| "max_allocated": 10993.732608 | |
| }, | |
| "latency": { | |
| "unit": "s", | |
| "mean": 28.168144866943358, | |
| "stdev": 0, | |
| "values": [ | |
| 28.168144866943358 | |
| ] | |
| }, | |
| "throughput": { | |
| "unit": "tokens/s", | |
| "value": 70.29216902117055 | |
| }, | |
| "energy": null, | |
| "efficiency": null | |
| }, | |
| "per_token": { | |
| "memory": null, | |
| "latency": { | |
| "unit": "s", | |
| "mean": 0.28452671582771066, | |
| "stdev": 0.0049875892887557205, | |
| "values": [ | |
| 0.2992322692871094, | |
| 0.2963630065917969, | |
| 0.2955704345703125, | |
| 0.29482598876953126, | |
| 0.29479730224609374, | |
| 0.29364532470703125, | |
| 0.2929387512207031, | |
| 0.29414093017578125, | |
| 0.29279232788085935, | |
| 0.2917713928222656, | |
| 0.29185842895507813, | |
| 0.2903203735351563, | |
| 0.2909624328613281, | |
| 0.29145086669921877, | |
| 0.2901166076660156, | |
| 0.2893209533691406, | |
| 0.29001214599609376, | |
| 0.29043405151367185, | |
| 0.28881613159179687, | |
| 0.28931686401367185, | |
| 0.28846795654296875, | |
| 0.28854989624023436, | |
| 0.2885509033203125, | |
| 0.28781976318359376, | |
| 0.28798974609375, | |
| 0.28707839965820314, | |
| 0.28717465209960935, | |
| 0.2871705627441406, | |
| 0.28670053100585935, | |
| 0.2863319091796875, | |
| 0.28658892822265625, | |
| 0.28629605102539063, | |
| 0.28560385131835936, | |
| 0.28722994995117185, | |
| 0.2855628662109375, | |
| 0.28619161987304687, | |
| 0.2864005126953125, | |
| 0.28537957763671873, | |
| 0.2857574462890625, | |
| 0.2856570739746094, | |
| 0.28499456787109373, | |
| 0.2846719970703125, | |
| 0.28515225219726564, | |
| 0.2836234130859375, | |
| 0.2847836303710938, | |
| 0.2847744140625, | |
| 0.2835599365234375, | |
| 0.28438424682617186, | |
| 0.28307763671875, | |
| 0.28367974853515626, | |
| 0.2833827819824219, | |
| 0.28306533813476564, | |
| 0.28318310546875, | |
| 0.283335693359375, | |
| 0.28157952880859377, | |
| 0.2825861206054687, | |
| 0.28271923828125, | |
| 0.2822932434082031, | |
| 0.28188568115234375, | |
| 0.2817966003417969, | |
| 0.2816174011230469, | |
| 0.2814392395019531, | |
| 0.28117196655273435, | |
| 0.28046029663085936, | |
| 0.2814535827636719, | |
| 0.28160614013671875, | |
| 0.2794721374511719, | |
| 0.2817894287109375, | |
| 0.2799134826660156, | |
| 0.28016229248046876, | |
| 0.27944244384765626, | |
| 0.28090060424804686, | |
| 0.27975372314453123, | |
| 0.2797445068359375, | |
| 0.28207308959960936, | |
| 0.2814228515625, | |
| 0.2799288330078125, | |
| 0.279947265625, | |
| 0.27979571533203124, | |
| 0.280489990234375, | |
| 0.27977828979492186, | |
| 0.2800814208984375, | |
| 0.27924581909179685, | |
| 0.28012442016601563, | |
| 0.2795233154296875, | |
| 0.2790953063964844, | |
| 0.2790983581542969, | |
| 0.2792755126953125, | |
| 0.278697998046875, | |
| 0.2785955810546875, | |
| 0.27879833984375, | |
| 0.2794700927734375, | |
| 0.2787348327636719, | |
| 0.2788065185546875, | |
| 0.27856280517578125, | |
| 0.2788147277832031, | |
| 0.27881063842773435, | |
| 0.2785607604980469, | |
| 0.2777671813964844 | |
| ] | |
| }, | |
| "throughput": { | |
| "unit": "tokens/s", | |
| "value": 70.29216902117055 | |
| }, | |
| "energy": null, | |
| "efficiency": null | |
| } | |
| } |