-
Notifications
You must be signed in to change notification settings - Fork 0
/
benchmark_temperature_14.txt
1313 lines (1249 loc) · 62.6 KB
/
benchmark_temperature_14.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000001:00:00.0
Warning: persistence mode is disabled on device 00000001:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000002:00:00.0
Warning: persistence mode is disabled on device 00000002:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000003:00:00.0
Warning: persistence mode is disabled on device 00000003:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000008:00:00.0
Warning: persistence mode is disabled on device 00000008:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000009:00:00.0
Warning: persistence mode is disabled on device 00000009:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000A:00:00.0
Warning: persistence mode is disabled on device 0000000A:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000B:00:00.0
Warning: persistence mode is disabled on device 0000000B:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000C:00:00.0
Warning: persistence mode is disabled on device 0000000C:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
All done.
All done.
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000001:00:00.0
Warning: persistence mode is disabled on device 00000001:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000002:00:00.0
Warning: persistence mode is disabled on device 00000002:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000003:00:00.0
Warning: persistence mode is disabled on device 00000003:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000008:00:00.0
Warning: persistence mode is disabled on device 00000008:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000009:00:00.0
Warning: persistence mode is disabled on device 00000009:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000A:00:00.0
Warning: persistence mode is disabled on device 0000000A:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000B:00:00.0
Warning: persistence mode is disabled on device 0000000B:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000C:00:00.0
Warning: persistence mode is disabled on device 0000000C:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
All done.
All done.
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000001:00:00.0
Warning: persistence mode is disabled on device 00000001:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000002:00:00.0
Warning: persistence mode is disabled on device 00000002:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000003:00:00.0
Warning: persistence mode is disabled on device 00000003:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000008:00:00.0
Warning: persistence mode is disabled on device 00000008:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000009:00:00.0
Warning: persistence mode is disabled on device 00000009:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000A:00:00.0
Warning: persistence mode is disabled on device 0000000A:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000B:00:00.0
Warning: persistence mode is disabled on device 0000000B:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000C:00:00.0
Warning: persistence mode is disabled on device 0000000C:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
All done.
All done.
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
Open ttft file
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 0.5576404333114624
P50 ttft = 0.45336318016052246
P99 ttft = 1.0570557188987733
Average tbt = 0.1607878108819326
P50 tbt = 0.020938789844512938
P99 tbt = 0.6155463654994968
All GPUs and memories are cold after 26.145309925079346
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 0.6646613734109061
P50 ttft = 0.5364360809326172
P99 ttft = 1.3857017517089845
Average tbt = 0.35721743674505335
P50 tbt = 0.3295293092727662
P99 tbt = 1.1996146535873418
All GPUs and memories are cold after 51.221529722213745
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 0.9690986973898751
P50 ttft = 0.7979626655578613
P99 ttft = 2.019521207809447
Average tbt = 0.9459668786185129
P50 tbt = 0.6404732942581178
P99 tbt = 3.649810641288757
All GPUs and memories are cold after 75.1019434928894
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 1.1748736079146223
P50 ttft = 0.8951623439788818
P99 ttft = 2.8861601829528816
Average tbt = 1.9618709988710357
P50 tbt = 1.718736004829407
P99 tbt = 5.317417283058168
All GPUs and memories are cold after 71.18941378593445
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 2.537920618057251
P50 ttft = 2.6023449897766113
P99 ttft = 4.460206573009491
Average tbt = 7.717322053909302
P50 tbt = 7.709369599819185
P99 tbt = 15.259901088953022
All GPUs and memories are cold after 62.09913086891174
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 4.806117292493582
P50 ttft = 5.232203960418701
P99 ttft = 9.197574608325958
Average tbt = 9.888317139446738
P50 tbt = 9.849328660964968
P99 tbt = 19.555668792009357
All GPUs and memories are cold after 79.11357164382935
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 7.020176008956073
P50 ttft = 7.351557493209839
P99 ttft = 13.543337812423706
Average tbt = 11.321179993185275
P50 tbt = 11.265444993972782
P99 tbt = 22.401435317039496
All GPUs and memories are cold after 80.11501359939575
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 9.133967112345868
P50 ttft = 8.927875757217407
P99 ttft = 17.462600321769713
Average tbt = 12.903249310585393
P50 tbt = 12.842646527290349
P99 tbt = 25.458675642013553
All GPUs and memories are cold after 65.10017132759094
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 0.5525533755620321
P50 ttft = 0.45219457149505615
P99 ttft = 1.0507950496673586
Average tbt = 0.14754553238550824
P50 tbt = 0.018778288364410402
P99 tbt = 0.6046465997695926
All GPUs and memories are cold after 16.119518518447876
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 0.6636845497857957
P50 ttft = 0.5340218544006348
P99 ttft = 1.3818969726562502
Average tbt = 0.3562290464128768
P50 tbt = 0.32857544422149665
P99 tbt = 1.196956462860108
All GPUs and memories are cold after 30.12201952934265
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 0.9668078013828822
P50 ttft = 0.7963030338287354
P99 ttft = 2.0114335775375354
Average tbt = 0.94283557006291
P50 tbt = 0.6382930040359498
P99 tbt = 3.6424251341819764
All GPUs and memories are cold after 42.13092660903931
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 1.1701628696627733
P50 ttft = 0.8948822021484375
P99 ttft = 2.877373981475831
Average tbt = 1.9580426303351801
P50 tbt = 1.7133507966995243
P99 tbt = 5.307908000946046
All GPUs and memories are cold after 46.109288692474365
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 2.5041086149215697
P50 ttft = 2.582106351852417
P99 ttft = 4.393991062641144
Average tbt = 7.656619493484499
P50 tbt = 7.650489258766176
P99 tbt = 15.177058883905413
All GPUs and memories are cold after 66.09065008163452
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 4.757323153316975
P50 ttft = 5.167678117752075
P99 ttft = 9.11738210916519
Average tbt = 9.858330714330082
P50 tbt = 9.829063451290132
P99 tbt = 19.493942791223528
All GPUs and memories are cold after 65.09144330024719
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 6.910615421321294
P50 ttft = 7.213823318481445
P99 ttft = 13.379851989746093
Average tbt = 11.283118118651931
P50 tbt = 11.242981839179995
P99 tbt = 22.280247300148016
All GPUs and memories are cold after 64.1213276386261
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 9.107502828161401
P50 ttft = 8.877403497695923
P99 ttft = 17.413480448722837
Average tbt = 12.88668845188187
P50 tbt = 12.850821352005008
P99 tbt = 25.419743920803075
All GPUs and memories are cold after 62.08594751358032
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 0.5505902568499247
P50 ttft = 0.4506704807281494
P99 ttft = 1.0475572586059572
Average tbt = 0.14737722675005596
P50 tbt = 0.01857062578201294
P99 tbt = 0.6038889646530156
All GPUs and memories are cold after 23.066954612731934
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 0.6639685517265683
P50 ttft = 0.5368664264678955
P99 ttft = 1.382395648956299
Average tbt = 0.3566013495127361
P50 tbt = 0.3290692329406739
P99 tbt = 1.1972587347030645
All GPUs and memories are cold after 39.11117482185364
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 0.9652665751320976
P50 ttft = 0.7959086894989014
P99 ttft = 2.005370473861693
Average tbt = 0.9427072218486245
P50 tbt = 0.6380706310272218
P99 tbt = 3.6376731219291685
All GPUs and memories are cold after 48.072622299194336
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 1.1732404639081258
P50 ttft = 0.8950474262237549
P99 ttft = 2.8781173706054695
Average tbt = 1.975903126088584
P50 tbt = 1.7468078136444094
P99 tbt = 5.3447268486022965
All GPUs and memories are cold after 57.10059094429016
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 2.392843179702759
P50 ttft = 2.462224245071411
P99 ttft = 4.290298302173614
Average tbt = 7.652878285408021
P50 tbt = 7.651297199726106
P99 tbt = 15.098855649471286
All GPUs and memories are cold after 57.07674789428711
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 4.7669075056910515
P50 ttft = 5.176513671875
P99 ttft = 9.134917690753936
Average tbt = 9.862061945348982
P50 tbt = 9.838299524784091
P99 tbt = 19.500126389026647
All GPUs and memories are cold after 71.09879899024963
92886
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000001:00:00.0
Warning: persistence mode is disabled on device 00000001:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000002:00:00.0
Warning: persistence mode is disabled on device 00000002:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000003:00:00.0
Warning: persistence mode is disabled on device 00000003:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000008:00:00.0
Warning: persistence mode is disabled on device 00000008:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000009:00:00.0
Warning: persistence mode is disabled on device 00000009:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000A:00:00.0
Warning: persistence mode is disabled on device 0000000A:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000B:00:00.0
Warning: persistence mode is disabled on device 0000000B:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000C:00:00.0
Warning: persistence mode is disabled on device 0000000C:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
All done.
All done.
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000001:00:00.0
Warning: persistence mode is disabled on device 00000001:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000002:00:00.0
Warning: persistence mode is disabled on device 00000002:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000003:00:00.0
Warning: persistence mode is disabled on device 00000003:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000008:00:00.0
Warning: persistence mode is disabled on device 00000008:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000009:00:00.0
Warning: persistence mode is disabled on device 00000009:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000A:00:00.0
Warning: persistence mode is disabled on device 0000000A:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000B:00:00.0
Warning: persistence mode is disabled on device 0000000B:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000C:00:00.0
Warning: persistence mode is disabled on device 0000000C:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
All done.
All done.
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000001:00:00.0
Warning: persistence mode is disabled on device 00000001:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000002:00:00.0
Warning: persistence mode is disabled on device 00000002:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000003:00:00.0
Warning: persistence mode is disabled on device 00000003:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000008:00:00.0
Warning: persistence mode is disabled on device 00000008:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000009:00:00.0
Warning: persistence mode is disabled on device 00000009:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000A:00:00.0
Warning: persistence mode is disabled on device 0000000A:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000B:00:00.0
Warning: persistence mode is disabled on device 0000000B:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000C:00:00.0
Warning: persistence mode is disabled on device 0000000C:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
All done.
All done.
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 6.9284821242502295
P50 ttft = 7.216264486312866
P99 ttft = 13.415212936401367
Average tbt = 11.294053480069929
P50 tbt = 11.266337060928347
P99 tbt = 22.307548652648933
All GPUs and memories are cold after 80.10934257507324
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 8.999272406819355
P50 ttft = 8.71572208404541
P99 ttft = 17.24712297439575
Average tbt = 12.84472349350711
P50 tbt = 12.844265270233159
P99 tbt = 25.307629974842072
All GPUs and memories are cold after 79.11095809936523
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 0.5522560079892477
P50 ttft = 0.4510679244995117
P99 ttft = 1.0507166957855225
Average tbt = 0.1477215886116028
P50 tbt = 0.018720376491546634
P99 tbt = 0.6053004014492038
All GPUs and memories are cold after 30.073236227035522
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 0.6641973654429117
P50 ttft = 0.5352828502655029
P99 ttft = 1.3835256576538086
Average tbt = 0.35661022776649115
P50 tbt = 0.32883033752441415
P99 tbt = 1.1977962589263922
All GPUs and memories are cold after 44.098838090896606
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 0.9673436028616769
P50 ttft = 0.7957701683044434
P99 ttft = 2.0117365646362293
Average tbt = 0.9431056649344309
P50 tbt = 0.6371329545974733
P99 tbt = 3.6419751191139222
All GPUs and memories are cold after 53.11147689819336
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 1.1720607804089058
P50 ttft = 0.8953456878662109
P99 ttft = 2.8778100013732915
Average tbt = 1.9602837981247323
P50 tbt = 1.717142987251282
P99 tbt = 5.3080703449249285
All GPUs and memories are cold after 58.082524061203
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 2.490982632637024
P50 ttft = 2.5613831281661987
P99 ttft = 4.420655260086059
Average tbt = 7.711444264888764
P50 tbt = 7.668738579750063
P99 tbt = 15.225322945594792
All GPUs and memories are cold after 61.12144446372986
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 4.72398017719388
P50 ttft = 5.055891633033752
P99 ttft = 9.134704344272613
Average tbt = 9.90119707994163
P50 tbt = 9.930089151859285
P99 tbt = 19.509784919500355
All GPUs and memories are cold after 65.09122395515442
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 6.842110728564328
P50 ttft = 7.09082818031311
P99 ttft = 13.26941252708435
Average tbt = 11.253030989921257
P50 tbt = 11.240942692756656
P99 tbt = 22.20727819538117
All GPUs and memories are cold after 72.10728979110718
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 9.114372922713498
P50 ttft = 8.845185995101929
P99 ttft = 17.37667258262634
Average tbt = 12.854854333831605
P50 tbt = 12.836710596084597
P99 tbt = 25.403062561511998
All GPUs and memories are cold after 80.15578103065491
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 0.551570475101471
P50 ttft = 0.4520944356918335
P99 ttft = 1.0509623956680298
Average tbt = 0.14740002353986106
P50 tbt = 0.018563294410705568
P99 tbt = 0.6041005897521976
All GPUs and memories are cold after 26.055431127548218
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
8170 - 5
Average ttft = 0.6791231859297979
P50 ttft = 0.5323312282562256
P99 ttft = 1.494096565246582
Average tbt = 0.36370862665630516
P50 tbt = 0.32848577499389653
P99 tbt = 1.2756068658828739
All GPUs and memories are cold after 38.13223958015442
92886
288 - 96
288 - 256
288 - 1024
1053 - 96
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600