deepspaarse 17

AMD Ryzen 9 7950X 16-Core testing with a ASUS ROG STRIX X670E-E GAMING WIFI (1905 BIOS) and NVIDIA GeForce RTX 3080 10GB on Ubuntu 23.10 via the Phoronix Test Suite.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2403151-PTS-DEEPSPAA58
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results
Show Result Confidence Charts

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
a
March 15
  1 Hour, 29 Minutes
b
March 15
  30 Minutes
c
March 15
  30 Minutes
d
March 15
  30 Minutes
e
March 15
  1 Hour, 33 Minutes
Invert Hiding All Results Option
  54 Minutes

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):


deepspaarse 17OpenBenchmarking.orgPhoronix Test SuiteAMD Ryzen 9 7950X 16-Core @ 5.88GHz (16 Cores / 32 Threads)ASUS ROG STRIX X670E-E GAMING WIFI (1905 BIOS)AMD Device 14d82 x 16GB DRAM-6000MT/s G Skill F5-6000J3038F16G2000GB Samsung SSD 980 PRO 2TB + Western Digital WD_BLACK SN850X 2000GBNVIDIA GeForce RTX 3080 10GBNVIDIA GA102 HD AudioDELL U2723QEIntel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411Ubuntu 23.106.7.0-060700-generic (x86_64)GNOME Shell 45.2X Server 1.21.1.7NVIDIA 550.54.144.6.0OpenCL 3.0 CUDA 12.4.89GCC 13.2.0ext43840x2160ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLOpenCLCompilerFile-SystemScreen ResolutionDeepspaarse 17 BenchmarksSystem Logs- Transparent Huge Pages: madvise- Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xa601206 - Python 3.11.6- gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of Safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

abcdeResult OverviewPhoronix Test Suite100%101%101%102%102%Neural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseN.T.C.D.m - S.S.SN.T.C.D.m - S.S.SN.D.C.o.b.u.o.I - A.M.SR.5.S.I - S.S.SR.5.S.I - S.S.SN.D.C.o.b.u.o.I - A.M.SR.5.S.I - A.M.SR.5.S.I - A.M.SC.C.R.5.I - A.M.SC.C.R.5.I - A.M.SN.T.C.B.b.u.S.S.I - A.M.SN.T.C.B.b.u.S.S.I - A.M.SC.C.R.5.I - S.S.SC.C.R.5.I - S.S.SN.T.C.D.m - A.M.SN.T.C.D.m - A.M.SN.T.C.B.b.u.c - A.M.SN.T.C.B.b.u.c - A.M.SN.D.C.o.b.u.o.I - S.S.SN.D.C.o.b.u.o.I - S.S.SL.C.7.Q - A.M.SL.C.7.Q - A.M.SB.L.N.Q.A.S.I - A.M.SB.L.N.Q.A.S.I - A.M.SN.T.C.B.b.u.c - S.S.SN.T.C.B.b.u.c - S.S.SB.L.N.Q.A.S.I - S.S.SB.L.N.Q.A.S.I - S.S.SC.D.Y.C.S.I - A.M.SN.T.C.B.b.u.S.S.I - S.S.SN.T.C.B.b.u.S.S.I - S.S.SC.D.Y.C.S.I - A.M.SC.D.Y.C.S.I - S.S.SC.D.Y.C.S.I - S.S.SC.S.9.P.Y.P - S.S.SC.S.9.P.Y.P - A.M.SC.S.9.P.Y.P - A.M.SC.S.9.P.Y.P - S.S.SR.5.B - S.S.SR.5.B - S.S.SR.5.B - A.M.SR.5.B - A.M.SL.C.7.Q - S.S.SL.C.7.Q - S.S.S

deepspaarse 17deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Streamdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Streamdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Streamdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Streamdeepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Synchronous Single-Streamdeepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Synchronous Single-Streamdeepsparse: ResNet-50, Baseline - Asynchronous Multi-Streamdeepsparse: ResNet-50, Baseline - Asynchronous Multi-Streamdeepsparse: ResNet-50, Baseline - Synchronous Single-Streamdeepsparse: ResNet-50, Baseline - Synchronous Single-Streamdeepsparse: ResNet-50, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: ResNet-50, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: ResNet-50, Sparse INT8 - Synchronous Single-Streamdeepsparse: ResNet-50, Sparse INT8 - Synchronous Single-Streamdeepsparse: Llama2 Chat 7b Quantized - Asynchronous Multi-Streamdeepsparse: Llama2 Chat 7b Quantized - Asynchronous Multi-Streamdeepsparse: Llama2 Chat 7b Quantized - Synchronous Single-Streamdeepsparse: Llama2 Chat 7b Quantized - Synchronous Single-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Streamdeepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Synchronous Single-Streamdeepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Synchronous Single-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Streamdeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Streamdeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Streamdeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Synchronous Single-Streamdeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Synchronous Single-Streamdeepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Synchronous Single-Streamdeepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Synchronous Single-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Streamabcde21.5986369.726918.812453.1496929.85268.5925300.87413.3206280.616428.4928195.29055.11492379.95133.35181437.55230.69374.09591900.47957.6108131.3760279.443028.6159194.42945.1378126.339963.268199.003210.0959191.702341.7072117.66438.494237.6522212.447931.573331.6588430.425418.5738112.86558.852121.5097371.570218.598153.761021.2505376.363918.656253.5945928.16578.6079301.23533.3168280.427728.5051195.96015.09762376.57193.35591438.2710.69324.10981894.22037.6089131.409280.454928.5076195.40385.112126.676363.122799.49710.046192.50641.53116.34928.590237.7759211.751231.661431.5706431.811418.5138112.6498.869621.4734370.827518.540353.929821.2799375.865918.691353.494927.6098.6128300.11383.3287281.026728.458195.50055.10952385.7693.34251444.83330.68994.08531905.41567.606131.4587280.9828.4559196.33575.088127.025162.946699.407310.0546192.607641.5244116.56398.574337.7639211.81931.657731.5743430.888618.5539112.25238.900421.3153374.077518.614553.714521.2445376.522418.715653.4247920.27748.6817301.70053.3113280.448628.5145196.02375.09632381.27163.34881420.74450.70194.07911908.60787.5982131.5957280.510428.5083196.37275.0871126.411263.260899.498410.046192.305441.5819115.25668.671537.7156212.090431.701731.5308430.663718.5638112.56918.874821.4523371.94818.653553.602421.3119374.965318.664853.5694926.39138.6244301.16893.3176280.535328.5061195.74735.10302359.66053.38001433.24260.69574.09621900.56417.6006131.5494282.484828.3070195.93555.0981126.725863.096699.104010.0857190.791641.9183117.20798.526937.6057212.710331.559131.6738429.152918.6294112.53448.878021.4247372.492318.616853.7068OpenBenchmarking.org

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Streamabcde510152025SE +/- 0.10, N = 3SE +/- 0.03, N = 321.6021.2521.2821.2421.31

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Streamabcde80160240320400SE +/- 1.41, N = 3SE +/- 0.28, N = 3369.73376.36375.87376.52374.97

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Streamabcde510152025SE +/- 0.03, N = 3SE +/- 0.02, N = 318.8118.6618.6918.7218.66

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Streamabcde1224364860SE +/- 0.08, N = 3SE +/- 0.06, N = 353.1553.5953.4953.4253.57

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Streamabcde2004006008001000SE +/- 1.67, N = 3SE +/- 1.13, N = 3929.85928.17927.61920.28926.39

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Streamabcde246810SE +/- 0.0153, N = 3SE +/- 0.0104, N = 38.59258.60798.61288.68178.6244

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Synchronous Single-Streamabcde70140210280350SE +/- 0.77, N = 3SE +/- 1.34, N = 3300.87301.24300.11301.70301.17

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Synchronous Single-Streamabcde0.7491.4982.2472.9963.745SE +/- 0.0086, N = 3SE +/- 0.0148, N = 33.32063.31683.32873.31133.3176

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Streamabcde60120180240300SE +/- 0.32, N = 3SE +/- 0.54, N = 3280.62280.43281.03280.45280.54

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Streamabcde714212835SE +/- 0.03, N = 3SE +/- 0.05, N = 328.4928.5128.4628.5128.51

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Baseline - Scenario: Synchronous Single-Streamabcde4080120160200SE +/- 0.13, N = 3SE +/- 0.46, N = 3195.29195.96195.50196.02195.75

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Baseline - Scenario: Synchronous Single-Streamabcde1.15092.30183.45274.60365.7545SE +/- 0.0035, N = 3SE +/- 0.0121, N = 35.11495.09765.10955.09635.1030

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Streamabcde5001000150020002500SE +/- 2.33, N = 3SE +/- 13.77, N = 32379.952376.572385.772381.272359.66

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Streamabcde0.76051.5212.28153.0423.8025SE +/- 0.0036, N = 3SE +/- 0.0197, N = 33.35183.35593.34253.34883.3800

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-Streamabcde30060090012001500SE +/- 2.00, N = 3SE +/- 6.10, N = 31437.551438.271444.831420.741433.24

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-Streamabcde0.15790.31580.47370.63160.7895SE +/- 0.0010, N = 3SE +/- 0.0030, N = 30.69370.69320.68990.70190.6957

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: Llama2 Chat 7b Quantized - Scenario: Asynchronous Multi-Streamabcde0.92471.84942.77413.69884.6235SE +/- 0.0014, N = 3SE +/- 0.0043, N = 34.09594.10984.08534.07914.0962

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: Llama2 Chat 7b Quantized - Scenario: Asynchronous Multi-Streamabcde400800120016002000SE +/- 0.71, N = 3SE +/- 1.74, N = 31900.481894.221905.421908.611900.56

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: Llama2 Chat 7b Quantized - Scenario: Synchronous Single-Streamabcde246810SE +/- 0.0050, N = 3SE +/- 0.0028, N = 37.61087.60897.60607.59827.6006

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: Llama2 Chat 7b Quantized - Scenario: Synchronous Single-Streamabcde306090120150SE +/- 0.09, N = 3SE +/- 0.05, N = 3131.38131.41131.46131.60131.55

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Streamabcde60120180240300SE +/- 0.35, N = 3SE +/- 0.79, N = 3279.44280.45280.98280.51282.48

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Streamabcde714212835SE +/- 0.04, N = 3SE +/- 0.08, N = 328.6228.5128.4628.5128.31

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Streamabcde4080120160200SE +/- 0.40, N = 3SE +/- 0.48, N = 3194.43195.40196.34196.37195.94

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Streamabcde1.1562.3123.4684.6245.78SE +/- 0.0107, N = 3SE +/- 0.0126, N = 35.13785.11205.08805.08715.0981

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Streamabcde306090120150SE +/- 0.41, N = 3SE +/- 0.88, N = 3126.34126.68127.03126.41126.73

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Streamabcde1428425670SE +/- 0.21, N = 3SE +/- 0.43, N = 363.2763.1262.9563.2663.10

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-Streamabcde20406080100SE +/- 0.04, N = 3SE +/- 0.14, N = 399.0099.5099.4199.5099.10

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-Streamabcde3691215SE +/- 0.00, N = 3SE +/- 0.01, N = 310.1010.0510.0510.0510.09

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Streamabcde4080120160200SE +/- 0.07, N = 3SE +/- 0.41, N = 3191.70192.51192.61192.31190.79

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Streamabcde1020304050SE +/- 0.02, N = 3SE +/- 0.09, N = 341.7141.5341.5241.5841.92

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Streamabcde306090120150SE +/- 0.36, N = 3SE +/- 0.10, N = 3117.66116.35116.56115.26117.21

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Streamabcde246810SE +/- 0.0262, N = 3SE +/- 0.0075, N = 38.49428.59028.57438.67158.5269

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Streamabcde918273645SE +/- 0.04, N = 3SE +/- 0.02, N = 337.6537.7837.7637.7237.61

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Streamabcde50100150200250SE +/- 0.21, N = 3SE +/- 0.13, N = 3212.45211.75211.82212.09212.71

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Streamabcde714212835SE +/- 0.01, N = 3SE +/- 0.02, N = 331.5731.6631.6631.7031.56

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Streamabcde714212835SE +/- 0.01, N = 3SE +/- 0.02, N = 331.6631.5731.5731.5331.67

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Streamabcde90180270360450SE +/- 0.23, N = 3SE +/- 0.70, N = 3430.43431.81430.89430.66429.15

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Streamabcde510152025SE +/- 0.01, N = 3SE +/- 0.03, N = 318.5718.5118.5518.5618.63

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Synchronous Single-Streamabcde306090120150SE +/- 0.54, N = 3SE +/- 0.18, N = 3112.87112.65112.25112.57112.53

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Synchronous Single-Streamabcde246810SE +/- 0.0421, N = 3SE +/- 0.0141, N = 38.85218.86968.90048.87488.8780

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Streamabcde510152025SE +/- 0.08, N = 3SE +/- 0.08, N = 321.5121.4721.3221.4521.42

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Streamabcde80160240320400SE +/- 1.38, N = 3SE +/- 1.05, N = 3371.57370.83374.08371.95372.49

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Streamabcde510152025SE +/- 0.01, N = 3SE +/- 0.03, N = 318.6018.5418.6118.6518.62

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Streamabcde1224364860SE +/- 0.02, N = 3SE +/- 0.08, N = 353.7653.9353.7153.6053.71

44 Results Shown

Neural Magic DeepSparse:
  NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Stream:
    items/sec
    ms/batch
  NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Stream:
    items/sec
    ms/batch
  NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Asynchronous Multi-Stream:
    items/sec
    ms/batch
  NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Synchronous Single-Stream:
    items/sec
    ms/batch
  ResNet-50, Baseline - Asynchronous Multi-Stream:
    items/sec
    ms/batch
  ResNet-50, Baseline - Synchronous Single-Stream:
    items/sec
    ms/batch
  ResNet-50, Sparse INT8 - Asynchronous Multi-Stream:
    items/sec
    ms/batch
  ResNet-50, Sparse INT8 - Synchronous Single-Stream:
    items/sec
    ms/batch
  Llama2 Chat 7b Quantized - Asynchronous Multi-Stream:
    items/sec
    ms/batch
  Llama2 Chat 7b Quantized - Synchronous Single-Stream:
    items/sec
    ms/batch
  CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Stream:
    items/sec
    ms/batch
  CV Classification, ResNet-50 ImageNet - Synchronous Single-Stream:
    items/sec
    ms/batch
  CV Detection, YOLOv5s COCO, Sparse INT8 - Asynchronous Multi-Stream:
    items/sec
    ms/batch
  CV Detection, YOLOv5s COCO, Sparse INT8 - Synchronous Single-Stream:
    items/sec
    ms/batch
  NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Stream:
    items/sec
    ms/batch
  NLP Text Classification, DistilBERT mnli - Synchronous Single-Stream:
    items/sec
    ms/batch
  CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Stream:
    items/sec
    ms/batch
  CV Segmentation, 90% Pruned YOLACT Pruned - Synchronous Single-Stream:
    items/sec
    ms/batch
  BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Stream:
    items/sec
    ms/batch
  BERT-Large, NLP Question Answering, Sparse INT8 - Synchronous Single-Stream:
    items/sec
    ms/batch
  NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Stream:
    items/sec
    ms/batch
  NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Stream:
    items/sec
    ms/batch