kreas commited on
Commit
bb02ed5
·
verified ·
1 Parent(s): 194e177

Upload /Llama-2-7b-hf/fp4_batch_size_1_sq_len_256_new_tokens_256/benchmark_report.json with huggingface_hub

Browse files
Llama-2-7b-hf/fp4_batch_size_1_sq_len_256_new_tokens_256/benchmark_report.json CHANGED
@@ -2,217 +2,210 @@
2
  "prefill": {
3
  "memory": {
4
  "unit": "MB",
5
- "max_ram": 1054.02368,
6
  "max_vram": 4940.890112,
7
  "max_reserved": 4456.448,
8
  "max_allocated": 4286.221824
9
  },
10
  "latency": {
11
  "unit": "s",
12
- "mean": 0.05123750924090948,
13
- "stdev": 0.0010706502176656841,
14
  "values": [
15
- 0.0659755859375,
16
- 0.05199052810668945,
17
- 0.051566593170166014,
18
- 0.05144793701171875,
19
- 0.051515392303466793,
20
- 0.051692543029785154,
21
- 0.05154099273681641,
22
- 0.05144063949584961,
23
- 0.05149593734741211,
24
- 0.05159731292724609,
25
- 0.05149491119384766,
26
- 0.05157068634033203,
27
- 0.05126553726196289,
28
- 0.05139353561401367,
29
- 0.05117747116088867,
30
- 0.05105868911743164,
31
- 0.05110067367553711,
32
- 0.051130367279052735,
33
- 0.051046398162841795,
34
- 0.051244033813476565,
35
- 0.050985984802246094,
36
- 0.05114470291137695,
37
- 0.05106790542602539,
38
- 0.05124505615234375,
39
- 0.05098188781738281,
40
- 0.051090431213378903,
41
- 0.05091123199462891,
42
- 0.05125120162963867,
43
- 0.0510648307800293,
44
- 0.051103870391845704,
45
- 0.05100953674316406,
46
- 0.05100339126586914,
47
- 0.05099126434326172,
48
- 0.05099008178710938,
49
- 0.050902015686035154,
50
- 0.051093505859375,
51
- 0.0510300178527832,
52
- 0.05105478286743164,
53
- 0.0510382080078125,
54
- 0.05108428955078125,
55
- 0.05131161499023437,
56
- 0.051297279357910154,
57
- 0.05113139343261719,
58
- 0.05104230499267578,
59
- 0.05096371078491211,
60
- 0.05103513717651367,
61
- 0.0509306869506836,
62
- 0.051179519653320314,
63
- 0.05130137634277344,
64
- 0.05116723251342774,
65
- 0.0511058235168457,
66
- 0.05115903854370117,
67
- 0.05106380844116211,
68
- 0.0511016960144043,
69
- 0.0510648307800293,
70
- 0.051130367279052735,
71
- 0.05098291015625,
72
- 0.0510648307800293,
73
- 0.05090508651733398,
74
- 0.05104230499267578,
75
- 0.05079859161376953,
76
- 0.050977790832519534,
77
- 0.050872318267822264,
78
- 0.05089484786987305,
79
- 0.05125222396850586,
80
- 0.05102796936035156,
81
- 0.05113446426391602,
82
- 0.051074047088623044,
83
- 0.05108633422851563,
84
- 0.05099008178710938,
85
- 0.05103411102294922,
86
- 0.05095030212402344,
87
- 0.05103529739379883,
88
- 0.05100543975830078,
89
- 0.05089382553100586,
90
- 0.050991104125976565,
91
- 0.050944286346435545,
92
- 0.05083238220214844,
93
- 0.0508671989440918,
94
- 0.05096371078491211,
95
- 0.05088256072998047,
96
- 0.050948097229003904,
97
- 0.05092761611938477,
98
- 0.050953216552734375,
99
- 0.05083238220214844,
100
- 0.05093273544311523,
101
- 0.05092147064208984,
102
- 0.05116118240356445,
103
- 0.051272705078125,
104
- 0.05120512008666992,
105
- 0.050945022583007815,
106
- 0.0511539192199707,
107
- 0.0510300178527832,
108
- 0.05109145736694336,
109
- 0.05091430282592774,
110
- 0.051037185668945315,
111
- 0.05093606567382813,
112
- 0.051130496978759765,
113
- 0.05101260757446289,
114
- 0.051171329498291014,
115
- 0.051135486602783206,
116
- 0.05121843338012695,
117
- 0.051056640625,
118
- 0.051106048583984376,
119
- 0.05103308868408203,
120
- 0.05125529479980469,
121
- 0.05099212646484375,
122
- 0.05107814407348633,
123
- 0.05135769653320312,
124
- 0.05131161499023437,
125
- 0.051108863830566405,
126
- 0.05118873596191406,
127
- 0.051081214904785156,
128
- 0.05114470291137695,
129
- 0.05098393630981445,
130
- 0.05115903854370117,
131
- 0.051130367279052735,
132
- 0.05137100982666016,
133
- 0.051094623565673826,
134
- 0.05132492828369141,
135
- 0.05114265441894531,
136
- 0.05147046279907227,
137
- 0.051356670379638675,
138
- 0.05125939178466797,
139
- 0.05102182388305664,
140
- 0.05111724853515625,
141
- 0.05101772689819336,
142
- 0.05130656051635742,
143
- 0.05165158462524414,
144
- 0.051326976776123044,
145
- 0.051297279357910154,
146
- 0.051323902130126955,
147
- 0.05111705780029297,
148
- 0.05118259048461914,
149
- 0.05120819091796875,
150
- 0.05118361663818359,
151
- 0.05120512008666992,
152
- 0.05110784149169922,
153
- 0.051320831298828126,
154
- 0.05113958358764648,
155
- 0.050966529846191405,
156
- 0.05113241577148438,
157
- 0.051153057098388674,
158
- 0.05125222396850586,
159
- 0.05120307159423828,
160
- 0.051158016204833984,
161
- 0.05113958358764648,
162
- 0.05109862518310547,
163
- 0.05151948928833008,
164
- 0.051320831298828126,
165
- 0.05114470291137695,
166
- 0.05106892776489258,
167
- 0.05124095916748047,
168
- 0.051149822235107424,
169
- 0.05108736038208008,
170
- 0.051214336395263675,
171
- 0.0511723518371582,
172
- 0.051310592651367185,
173
- 0.05104435348510742,
174
- 0.05119385528564453,
175
- 0.051253246307373046,
176
- 0.0513361930847168,
177
- 0.05121542358398438,
178
- 0.05108428955078125,
179
- 0.05111500930786133,
180
- 0.051211265563964846,
181
- 0.05112934494018555,
182
- 0.05112115097045898,
183
- 0.05118054580688477,
184
- 0.051609600067138675,
185
- 0.05122355270385742,
186
- 0.051326976776123044,
187
- 0.0512911376953125,
188
- 0.05137715148925781,
189
- 0.05134233474731445,
190
- 0.05137203216552735,
191
- 0.051113983154296876,
192
- 0.05134233474731445,
193
- 0.05124300765991211,
194
- 0.0512973747253418,
195
- 0.0511539192199707,
196
- 0.05131468963623047,
197
- 0.0512174072265625,
198
- 0.05145600128173828,
199
- 0.0512911376953125,
200
- 0.05134233474731445,
201
- 0.05123276901245117,
202
- 0.05135974502563476,
203
- 0.05123891067504883,
204
- 0.05128396987915039,
205
- 0.051138561248779295,
206
- 0.051160064697265625,
207
- 0.051156993865966796,
208
- 0.05127167892456055,
209
- 0.051125247955322264,
210
- 0.05130547332763672
211
  ]
212
  },
213
  "throughput": {
214
  "unit": "tokens/s",
215
- "value": 4996.339669759012
216
  },
217
  "energy": null,
218
  "efficiency": null
@@ -220,22 +213,22 @@
220
  "decode": {
221
  "memory": {
222
  "unit": "MB",
223
- "max_ram": 1073.946624,
224
  "max_vram": 5314.183168,
225
  "max_reserved": 4829.741056,
226
  "max_allocated": 4749.821952
227
  },
228
  "latency": {
229
  "unit": "s",
230
- "mean": 10.52300802230834,
231
  "stdev": 0,
232
  "values": [
233
- 10.52300802230834
234
  ]
235
  },
236
  "throughput": {
237
  "unit": "tokens/s",
238
- "value": 24.23261480552049
239
  },
240
  "energy": null,
241
  "efficiency": null
@@ -244,269 +237,269 @@
244
  "memory": null,
245
  "latency": {
246
  "unit": "s",
247
- "mean": 0.04126669812669938,
248
- "stdev": 0.00018803687120235083,
249
  "values": [
250
- 0.041062400817871096,
251
- 0.04218368148803711,
252
- 0.04143513488769531,
253
- 0.04141260910034179,
254
- 0.04122438430786133,
255
- 0.04182201766967773,
256
- 0.04131532669067383,
257
- 0.041508926391601565,
258
- 0.0412476806640625,
259
- 0.0414648323059082,
260
- 0.041390079498291016,
261
- 0.041376766204833985,
262
- 0.04126822280883789,
263
- 0.04216320037841797,
264
- 0.04141363143920898,
265
- 0.04158464050292969,
266
- 0.0414463996887207,
267
- 0.041447425842285154,
268
- 0.041365505218505856,
269
- 0.04155187225341797,
270
- 0.04128255844116211,
271
- 0.04147609710693359,
272
- 0.04125798416137695,
273
- 0.041586688995361325,
274
- 0.04130918502807617,
275
- 0.04144332885742188,
276
- 0.04130630493164063,
277
- 0.04154758453369141,
278
- 0.04120678329467774,
279
- 0.04144025421142578,
280
- 0.04155392074584961,
281
- 0.041575424194335936,
282
- 0.04120576095581055,
283
- 0.04138393783569336,
284
- 0.04116070556640625,
285
- 0.041366527557373044,
286
- 0.04109414291381836,
287
- 0.04132147216796875,
288
- 0.04137779235839844,
289
- 0.041406654357910154,
290
- 0.0411451530456543,
291
- 0.04208332824707031,
292
- 0.041381889343261716,
293
- 0.04139622497558594,
294
- 0.04112793731689453,
295
- 0.04124671936035156,
296
- 0.041082878112792966,
297
- 0.041289726257324216,
298
- 0.04118527984619141,
299
- 0.04127436828613281,
300
- 0.041280513763427736,
301
- 0.041289726257324216,
302
- 0.041140350341796875,
303
- 0.041253761291503904,
304
- 0.04111872100830078,
305
- 0.04134707260131836,
306
- 0.041063423156738284,
307
- 0.04120678329467774,
308
- 0.04097945785522461,
309
- 0.04124671936035156,
310
- 0.04109721755981445,
311
- 0.04122214508056641,
312
- 0.04116070556640625,
313
- 0.041442337036132815,
314
- 0.04120572662353516,
315
- 0.041280513763427736,
316
- 0.04173004913330078,
317
- 0.04132147216796875,
318
- 0.041417728424072264,
319
- 0.04125408172607422,
320
- 0.04129167938232422,
321
- 0.041525150299072264,
322
- 0.041381889343261716,
323
- 0.04137472152709961,
324
- 0.04124160003662109,
325
- 0.04126105499267578,
326
- 0.04111360168457031,
327
- 0.04107059097290039,
328
- 0.041022464752197264,
329
- 0.041221118927001955,
330
- 0.041191425323486325,
331
- 0.04099379348754883,
332
- 0.04117913436889648,
333
- 0.04115881729125977,
334
- 0.04118921661376953,
335
- 0.041196640014648435,
336
- 0.04112067031860352,
337
- 0.04149875259399414,
338
- 0.04127033615112305,
339
- 0.04110316848754883,
340
- 0.04128460693359375,
341
- 0.041215999603271485,
342
- 0.04124671936035156,
343
- 0.0420423698425293,
344
- 0.04139519882202149,
345
- 0.04117299270629883,
346
- 0.04117708969116211,
347
- 0.04122934341430664,
348
- 0.04120163345336914,
349
- 0.04123971176147461,
350
- 0.0410939826965332,
351
- 0.041296897888183595,
352
- 0.04132579040527344,
353
- 0.041259807586669923,
354
- 0.04108492660522461,
355
- 0.041302207946777345,
356
- 0.04109292984008789,
357
- 0.041171966552734376,
358
- 0.041603073120117184,
359
- 0.04114246368408203,
360
- 0.041107265472412106,
361
- 0.04137062454223633,
362
- 0.041234527587890625,
363
- 0.04107766342163086,
364
- 0.041234432220458986,
365
- 0.04126003265380859,
366
- 0.04125900650024414,
367
- 0.041169921875,
368
- 0.04117401504516602,
369
- 0.041027584075927735,
370
- 0.041105567932128904,
371
- 0.041274208068847656,
372
- 0.04106649780273437,
373
- 0.0410511360168457,
374
- 0.041181182861328124,
375
- 0.04137984085083008,
376
- 0.04116604614257813,
377
- 0.04192233657836914,
378
- 0.041285633087158206,
379
- 0.04117628860473633,
380
- 0.04104374313354492,
381
- 0.04105420684814453,
382
- 0.041073665618896485,
383
- 0.041078014373779295,
384
- 0.04116147232055664,
385
- 0.0412231674194336,
386
- 0.041221118927001955,
387
- 0.04113411331176758,
388
- 0.04107158279418945,
389
- 0.04097740936279297,
390
- 0.04116089630126953,
391
- 0.04098233413696289,
392
- 0.04112793731689453,
393
- 0.0411146240234375,
394
- 0.041134078979492186,
395
- 0.04099174499511719,
396
- 0.04132966232299805,
397
- 0.041134078979492186,
398
- 0.041591808319091796,
399
- 0.04124671936035156,
400
- 0.04135321426391601,
401
- 0.041616382598876955,
402
- 0.04127641677856445,
403
- 0.04111667251586914,
404
- 0.041171966552734376,
405
- 0.04110028839111328,
406
- 0.041215999603271485,
407
- 0.04108502578735351,
408
- 0.041145374298095706,
409
- 0.04119744110107422,
410
- 0.04119862365722656,
411
- 0.041180126190185545,
412
- 0.041275390625,
413
- 0.04118425750732422,
414
- 0.04125286483764649,
415
- 0.04107571029663086,
416
- 0.0412149772644043,
417
- 0.04124364852905273,
418
- 0.04109107208251953,
419
- 0.041201663970947267,
420
- 0.04115865707397461,
421
- 0.04196249771118164,
422
- 0.04115558242797852,
423
- 0.04115865707397461,
424
- 0.041148414611816404,
425
- 0.04124160003662109,
426
- 0.04115558242797852,
427
- 0.041234432220458986,
428
- 0.04120678329467774,
429
- 0.04126822280883789,
430
- 0.04114636611938476,
431
- 0.041220382690429686,
432
- 0.04111536026000977,
433
- 0.04098252868652344,
434
- 0.04126844787597656,
435
- 0.041220897674560546,
436
- 0.04111360168457031,
437
- 0.0410511360168457,
438
- 0.041191425323486325,
439
- 0.041060577392578124,
440
- 0.041229087829589846,
441
- 0.04119551849365234,
442
- 0.041219070434570314,
443
- 0.041605342864990236,
444
- 0.04123830413818359,
445
- 0.041371646881103515,
446
- 0.04118732833862305,
447
- 0.04122528076171875,
448
- 0.04120364761352539,
449
- 0.0412344970703125,
450
- 0.04112905502319336,
451
- 0.04119126510620117,
452
- 0.04107571029663086,
453
- 0.041158878326416015,
454
- 0.0411195182800293,
455
- 0.04115865707397461,
456
- 0.04118527984619141,
457
- 0.04110438537597656,
458
- 0.04120576095581055,
459
- 0.04112588882446289,
460
- 0.04110540771484375,
461
- 0.0411514892578125,
462
- 0.04126310348510742,
463
- 0.04111360168457031,
464
- 0.04151705551147461,
465
- 0.0413040657043457,
466
- 0.04134403228759766,
467
- 0.04123440170288086,
468
- 0.041365505218505856,
469
- 0.04127231979370117,
470
- 0.041270271301269534,
471
- 0.04121721649169922,
472
- 0.041123649597167966,
473
- 0.041204734802246096,
474
- 0.04134707260131836,
475
- 0.04141574478149414,
476
- 0.041328575134277346,
477
- 0.04124684906005859,
478
- 0.041325439453125,
479
- 0.04120678329467774,
480
- 0.04122521591186523,
481
- 0.04125183868408203,
482
- 0.04155187225341797,
483
- 0.04138598251342773,
484
- 0.041444351196289066,
485
- 0.04127231979370117,
486
- 0.041280513763427736,
487
- 0.04124265670776367,
488
- 0.04112380981445313,
489
- 0.04129792022705078,
490
- 0.041215999603271485,
491
- 0.041319423675537106,
492
- 0.04139929580688476,
493
- 0.041441280364990236,
494
- 0.04137267303466797,
495
- 0.041414657592773435,
496
- 0.041227390289306644,
497
- 0.04129267120361328,
498
- 0.04110233688354492,
499
- 0.04132556915283203,
500
- 0.041224193572998044,
501
- 0.04122726440429687,
502
- 0.041543678283691404,
503
- 0.0412231674194336,
504
- 0.04120064163208008
505
  ]
506
  },
507
  "throughput": {
508
  "unit": "tokens/s",
509
- "value": 24.23261480552049
510
  },
511
  "energy": null,
512
  "efficiency": null
 
2
  "prefill": {
3
  "memory": {
4
  "unit": "MB",
5
+ "max_ram": 1038.516224,
6
  "max_vram": 4940.890112,
7
  "max_reserved": 4456.448,
8
  "max_allocated": 4286.221824
9
  },
10
  "latency": {
11
  "unit": "s",
12
+ "mean": 0.052913485915572565,
13
+ "stdev": 0.00032505269904713316,
14
  "values": [
15
+ 0.056078624725341794,
16
+ 0.053402687072753904,
17
+ 0.05325516891479492,
18
+ 0.05332992172241211,
19
+ 0.05328793716430664,
20
+ 0.05345280075073242,
21
+ 0.05353779220581055,
22
+ 0.05339353561401367,
23
+ 0.05308415985107422,
24
+ 0.053292030334472655,
25
+ 0.05360639953613281,
26
+ 0.05315584182739258,
27
+ 0.05302579116821289,
28
+ 0.05301484680175781,
29
+ 0.053318656921386716,
30
+ 0.053171199798583986,
31
+ 0.052959232330322265,
32
+ 0.052942848205566405,
33
+ 0.052843582153320315,
34
+ 0.052894718170166014,
35
+ 0.0529090576171875,
36
+ 0.052863998413085936,
37
+ 0.052838401794433595,
38
+ 0.05293689727783203,
39
+ 0.05269708633422852,
40
+ 0.0528271369934082,
41
+ 0.05269404983520508,
42
+ 0.05289267349243164,
43
+ 0.05309439849853516,
44
+ 0.05286297607421875,
45
+ 0.052985855102539066,
46
+ 0.052980735778808595,
47
+ 0.05296646499633789,
48
+ 0.0530145263671875,
49
+ 0.052853759765625,
50
+ 0.052891647338867184,
51
+ 0.052706302642822264,
52
+ 0.05278617477416992,
53
+ 0.052757503509521485,
54
+ 0.052959232330322265,
55
+ 0.05290198516845703,
56
+ 0.052918270111083986,
57
+ 0.052847614288330076,
58
+ 0.0529090576171875,
59
+ 0.05295622253417969,
60
+ 0.05309235382080078,
61
+ 0.052929534912109374,
62
+ 0.053059585571289064,
63
+ 0.05308927917480469,
64
+ 0.05350502395629883,
65
+ 0.053425151824951174,
66
+ 0.05314559936523437,
67
+ 0.05325721740722656,
68
+ 0.05333401489257812,
69
+ 0.05292569732666016,
70
+ 0.05301248168945313,
71
+ 0.053029888153076174,
72
+ 0.053043201446533204,
73
+ 0.052752384185791014,
74
+ 0.0530247688293457,
75
+ 0.052754432678222656,
76
+ 0.05291417694091797,
77
+ 0.05271244812011719,
78
+ 0.05285683059692383,
79
+ 0.052673534393310545,
80
+ 0.052736000061035154,
81
+ 0.05307904052734375,
82
+ 0.05292342376708985,
83
+ 0.05288755035400391,
84
+ 0.052992000579833984,
85
+ 0.05299302291870117,
86
+ 0.052891647338867184,
87
+ 0.05287116622924805,
88
+ 0.05308620834350586,
89
+ 0.05290291213989258,
90
+ 0.05311897659301758,
91
+ 0.053116928100585936,
92
+ 0.05302387237548828,
93
+ 0.05302272033691406,
94
+ 0.053046272277832034,
95
+ 0.05314371109008789,
96
+ 0.05306470489501953,
97
+ 0.053028865814208986,
98
+ 0.053028865814208986,
99
+ 0.05291622543334961,
100
+ 0.05298175811767578,
101
+ 0.05284966278076172,
102
+ 0.052838401794433595,
103
+ 0.052779006958007815,
104
+ 0.05353574371337891,
105
+ 0.053269504547119144,
106
+ 0.052951038360595705,
107
+ 0.052945919036865234,
108
+ 0.05300428771972656,
109
+ 0.05282406234741211,
110
+ 0.05285068893432617,
111
+ 0.05282815933227539,
112
+ 0.05304729461669922,
113
+ 0.05282304000854492,
114
+ 0.052795391082763675,
115
+ 0.052424705505371094,
116
+ 0.05259980773925781,
117
+ 0.05231206512451172,
118
+ 0.052416511535644535,
119
+ 0.05259468841552734,
120
+ 0.052670463562011716,
121
+ 0.052517887115478515,
122
+ 0.052736000061035154,
123
+ 0.05315379333496094,
124
+ 0.05307904052734375,
125
+ 0.05294182586669922,
126
+ 0.05301760101318359,
127
+ 0.05291212844848633,
128
+ 0.05291622543334961,
129
+ 0.05297459030151367,
130
+ 0.05274828720092774,
131
+ 0.052923648834228516,
132
+ 0.053015552520751956,
133
+ 0.05287628936767578,
134
+ 0.05302067184448242,
135
+ 0.05293062210083008,
136
+ 0.0530032958984375,
137
+ 0.05308108901977539,
138
+ 0.05307392120361328,
139
+ 0.052964351654052735,
140
+ 0.05300227355957031,
141
+ 0.052999263763427736,
142
+ 0.05304143905639649,
143
+ 0.05324697494506836,
144
+ 0.05323878479003906,
145
+ 0.05293670272827149,
146
+ 0.052936767578125,
147
+ 0.052819969177246094,
148
+ 0.05294899368286133,
149
+ 0.05278236770629883,
150
+ 0.05269708633422852,
151
+ 0.05266124725341797,
152
+ 0.052721664428710936,
153
+ 0.05249228668212891,
154
+ 0.052691967010498046,
155
+ 0.05252000045776367,
156
+ 0.05259980773925781,
157
+ 0.052598785400390625,
158
+ 0.05289267349243164,
159
+ 0.05249433517456055,
160
+ 0.052751361846923826,
161
+ 0.052604927062988284,
162
+ 0.053031070709228516,
163
+ 0.05271756744384765,
164
+ 0.05282815933227539,
165
+ 0.052777088165283204,
166
+ 0.052625438690185544,
167
+ 0.0526869125366211,
168
+ 0.052670463562011716,
169
+ 0.0526110725402832,
170
+ 0.052615169525146485,
171
+ 0.05271142578125,
172
+ 0.05281484985351562,
173
+ 0.05275852966308594,
174
+ 0.05290086364746094,
175
+ 0.05276671981811523,
176
+ 0.05274009704589844,
177
+ 0.052708351135253906,
178
+ 0.052631553649902345,
179
+ 0.05263872146606445,
180
+ 0.052831230163574217,
181
+ 0.05260697555541992,
182
+ 0.052746238708496096,
183
+ 0.05256095886230469,
184
+ 0.05263974380493164,
185
+ 0.05304729461669922,
186
+ 0.05299302291870117,
187
+ 0.05269504165649414,
188
+ 0.052831230163574217,
189
+ 0.05263974380493164,
190
+ 0.05274521636962891,
191
+ 0.05263974380493164,
192
+ 0.052691967010498046,
193
+ 0.052732929229736325,
194
+ 0.05275872039794922,
195
+ 0.052596736907958984,
196
+ 0.05269606399536133,
197
+ 0.05270528030395508,
198
+ 0.052716545104980465,
199
+ 0.052686847686767575,
200
+ 0.052631553649902345,
201
+ 0.05256806564331055,
202
+ 0.05263052749633789,
203
+ 0.05260287857055664
 
 
 
 
 
 
 
204
  ]
205
  },
206
  "throughput": {
207
  "unit": "tokens/s",
208
+ "value": 4838.086086569068
209
  },
210
  "energy": null,
211
  "efficiency": null
 
213
  "decode": {
214
  "memory": {
215
  "unit": "MB",
216
+ "max_ram": 1058.439168,
217
  "max_vram": 5314.183168,
218
  "max_reserved": 4829.741056,
219
  "max_allocated": 4749.821952
220
  },
221
  "latency": {
222
  "unit": "s",
223
+ "mean": 10.808775684356693,
224
  "stdev": 0,
225
  "values": [
226
+ 10.808775684356693
227
  ]
228
  },
229
  "throughput": {
230
  "unit": "tokens/s",
231
+ "value": 23.59194116397993
232
  },
233
  "energy": null,
234
  "efficiency": null
 
237
  "memory": null,
238
  "latency": {
239
  "unit": "s",
240
+ "mean": 0.04238735562492821,
241
+ "stdev": 0.0001981619210019512,
242
  "values": [
243
+ 0.04212531280517578,
244
+ 0.04317900848388672,
245
+ 0.04262911987304688,
246
+ 0.04252876663208008,
247
+ 0.04242227172851563,
248
+ 0.04262297439575195,
249
+ 0.042409984588623044,
250
+ 0.04299264144897461,
251
+ 0.04246732711791992,
252
+ 0.04247552108764648,
253
+ 0.04245708847045898,
254
+ 0.04269772720336914,
255
+ 0.04220415878295898,
256
+ 0.042380287170410154,
257
+ 0.04230348968505859,
258
+ 0.042859519958496094,
259
+ 0.042592254638671875,
260
+ 0.042534912109375,
261
+ 0.042436607360839845,
262
+ 0.04259532928466797,
263
+ 0.042281982421875,
264
+ 0.04261478424072265,
265
+ 0.04221644973754883,
266
+ 0.042600448608398435,
267
+ 0.042548225402832034,
268
+ 0.042676223754882815,
269
+ 0.04239257431030274,
270
+ 0.04248371124267578,
271
+ 0.042365951538085936,
272
+ 0.04251443099975586,
273
+ 0.04271001434326172,
274
+ 0.04268134307861328,
275
+ 0.042347518920898435,
276
+ 0.042517505645751956,
277
+ 0.04224716949462891,
278
+ 0.04273049545288086,
279
+ 0.04242227172851563,
280
+ 0.04248166275024414,
281
+ 0.04232294464111328,
282
+ 0.042534912109375,
283
+ 0.04236492919921875,
284
+ 0.04240281677246094,
285
+ 0.04232294464111328,
286
+ 0.04239462280273437,
287
+ 0.042587135314941404,
288
+ 0.04257484817504883,
289
+ 0.04241100692749023,
290
+ 0.042630142211914065,
291
+ 0.04227993774414063,
292
+ 0.0428144645690918,
293
+ 0.04223385620117188,
294
+ 0.042484832763671876,
295
+ 0.04234230422973633,
296
+ 0.04235776138305664,
297
+ 0.042393600463867184,
298
+ 0.04244275283813476,
299
+ 0.04225228881835937,
300
+ 0.04241408157348633,
301
+ 0.042449920654296876,
302
+ 0.042550273895263675,
303
+ 0.04237004852294922,
304
+ 0.04248166275024414,
305
+ 0.04241100692749023,
306
+ 0.042635265350341796,
307
+ 0.04256256103515625,
308
+ 0.042446849822998046,
309
+ 0.0423454704284668,
310
+ 0.042188800811767575,
311
+ 0.042205184936523435,
312
+ 0.042120193481445314,
313
+ 0.042461185455322265,
314
+ 0.0427960319519043,
315
+ 0.04237619018554688,
316
+ 0.042469375610351565,
317
+ 0.04238950347900391,
318
+ 0.04258099365234375,
319
+ 0.042401790618896484,
320
+ 0.0423454704284668,
321
+ 0.04235878372192383,
322
+ 0.042308609008789064,
323
+ 0.04231679916381836,
324
+ 0.042240001678466796,
325
+ 0.04228403091430664,
326
+ 0.04237721633911133,
327
+ 0.042472446441650394,
328
+ 0.04232294464111328,
329
+ 0.04221952056884765,
330
+ 0.04295372772216797,
331
+ 0.04237209701538086,
332
+ 0.04271615982055664,
333
+ 0.04314931106567383,
334
+ 0.04329779052734375,
335
+ 0.04274995040893555,
336
+ 0.04253696060180664,
337
+ 0.04260659027099609,
338
+ 0.04251136016845703,
339
+ 0.042355712890625,
340
+ 0.04226355361938477,
341
+ 0.04225024032592774,
342
+ 0.04241100692749023,
343
+ 0.042308609008789064,
344
+ 0.04237721633911133,
345
+ 0.042352638244628905,
346
+ 0.04285542297363281,
347
+ 0.04242432022094727,
348
+ 0.04247449493408203,
349
+ 0.042365951538085936,
350
+ 0.042380287170410154,
351
+ 0.042470687866210936,
352
+ 0.0423851203918457,
353
+ 0.042466400146484375,
354
+ 0.04238838577270508,
355
+ 0.04231987380981445,
356
+ 0.042248191833496096,
357
+ 0.042264575958251956,
358
+ 0.042349567413330076,
359
+ 0.04227686309814453,
360
+ 0.04237004852294922,
361
+ 0.04231679916381836,
362
+ 0.04227993774414063,
363
+ 0.04231884765625,
364
+ 0.042587135314941404,
365
+ 0.042671104431152344,
366
+ 0.042380287170410154,
367
+ 0.04238950347900391,
368
+ 0.0425615348815918,
369
+ 0.04221542358398438,
370
+ 0.04232908630371094,
371
+ 0.04229632186889649,
372
+ 0.04226355361938477,
373
+ 0.04216729736328125,
374
+ 0.04230963134765625,
375
+ 0.042248191833496096,
376
+ 0.042336254119873046,
377
+ 0.04225228881835937,
378
+ 0.04223283386230469,
379
+ 0.04224716949462891,
380
+ 0.04240793609619141,
381
+ 0.04226662445068359,
382
+ 0.042231807708740236,
383
+ 0.042221569061279295,
384
+ 0.04224716949462891,
385
+ 0.04223897552490234,
386
+ 0.04257177734375,
387
+ 0.04232191848754883,
388
+ 0.04239257431030274,
389
+ 0.042395648956298826,
390
+ 0.04211916732788086,
391
+ 0.04237516784667969,
392
+ 0.0423004150390625,
393
+ 0.04225024032592774,
394
+ 0.04214476776123047,
395
+ 0.042011646270751955,
396
+ 0.04213862228393555,
397
+ 0.042106880187988284,
398
+ 0.042090496063232424,
399
+ 0.04208639907836914,
400
+ 0.04201779174804687,
401
+ 0.04231475067138672,
402
+ 0.042060798645019534,
403
+ 0.042109153747558595,
404
+ 0.042100513458251956,
405
+ 0.04212531280517578,
406
+ 0.04248166275024414,
407
+ 0.04213862228393555,
408
+ 0.04229119873046875,
409
+ 0.04220620727539062,
410
+ 0.042202110290527346,
411
+ 0.04238438415527344,
412
+ 0.04234854507446289,
413
+ 0.04245094299316406,
414
+ 0.04219801712036133,
415
+ 0.042657791137695314,
416
+ 0.04236083221435547,
417
+ 0.04247040176391602,
418
+ 0.042793983459472655,
419
+ 0.042499073028564455,
420
+ 0.042355712890625,
421
+ 0.04231679916381836,
422
+ 0.042246143341064454,
423
+ 0.04238950347900391,
424
+ 0.042444801330566405,
425
+ 0.042270721435546874,
426
+ 0.042142719268798826,
427
+ 0.042259456634521485,
428
+ 0.04219596862792969,
429
+ 0.04232294464111328,
430
+ 0.04226355361938477,
431
+ 0.04223283386230469,
432
+ 0.042687488555908204,
433
+ 0.042256385803222656,
434
+ 0.042447872161865234,
435
+ 0.042543102264404296,
436
+ 0.042359809875488284,
437
+ 0.0423372802734375,
438
+ 0.042343425750732425,
439
+ 0.042387454986572266,
440
+ 0.04230758285522461,
441
+ 0.04242739105224609,
442
+ 0.04239155197143555,
443
+ 0.04252364730834961,
444
+ 0.042426368713378904,
445
+ 0.04248086547851562,
446
+ 0.042447647094726565,
447
+ 0.04232396697998047,
448
+ 0.042829822540283204,
449
+ 0.04229119873046875,
450
+ 0.0423454704284668,
451
+ 0.04238848114013672,
452
+ 0.04245913696289062,
453
+ 0.042471424102783206,
454
+ 0.04237823867797851,
455
+ 0.04242943954467773,
456
+ 0.042856449127197264,
457
+ 0.042487808227539066,
458
+ 0.04250931167602539,
459
+ 0.04227481460571289,
460
+ 0.04231987380981445,
461
+ 0.04230553436279297,
462
+ 0.0424089584350586,
463
+ 0.042275840759277344,
464
+ 0.04236492919921875,
465
+ 0.042409984588623044,
466
+ 0.04235366439819336,
467
+ 0.04232499313354492,
468
+ 0.042280960083007815,
469
+ 0.042630142211914065,
470
+ 0.04246527862548828,
471
+ 0.04254515075683594,
472
+ 0.04259532928466797,
473
+ 0.04233932876586914,
474
+ 0.04229119873046875,
475
+ 0.042310657501220705,
476
+ 0.04226559829711914,
477
+ 0.04222259140014648,
478
+ 0.042259456634521485,
479
+ 0.04220723342895508,
480
+ 0.04205670547485352,
481
+ 0.04221235275268555,
482
+ 0.04204032135009766,
483
+ 0.04199935913085937,
484
+ 0.042223617553710936,
485
+ 0.04209766387939453,
486
+ 0.04221132659912109,
487
+ 0.042021888732910156,
488
+ 0.042608638763427735,
489
+ 0.042193920135498046,
490
+ 0.04217446517944336,
491
+ 0.04215500640869141,
492
+ 0.042156032562255856,
493
+ 0.0422031364440918,
494
+ 0.042231807708740236,
495
+ 0.04218982315063476,
496
+ 0.04215398406982422,
497
+ 0.04222873687744141
498
  ]
499
  },
500
  "throughput": {
501
  "unit": "tokens/s",
502
+ "value": 23.59194116397993
503
  },
504
  "energy": null,
505
  "efficiency": null