deepsparse 1.7 raptor lake

Intel Core i9-14900K testing with a ASUS PRIME Z790-P WIFI (1402 BIOS) and ASUS Intel RPL-S 31GB 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 2403152-PTS-DEEPSPAR00
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
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
a
March 15
  1 Hour, 30 Minutes
b
March 15
  30 Minutes
c
March 15
  30 Minutes
d
March 15
  1 Hour, 28 Minutes
e
March 15
  30 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):


deepsparse 1.7 raptor lakeOpenBenchmarking.orgPhoronix Test SuiteIntel Core i9-14900K @ 5.70GHz (24 Cores / 32 Threads)ASUS PRIME Z790-P WIFI (1402 BIOS)Intel Device 7a272 x 16GB DRAM-6000MT/s Corsair CMK32GX5M2B6000C36Western Digital WD_BLACK SN850X 1000GBASUS Intel RPL-S 31GB (1650MHz)Realtek ALC897ASUS VP28UUbuntu 23.106.8.0-phx (x86_64)GNOME Shell 45.1X Server 1.21.1.74.6 Mesa 24.0~git2312240600.c05261~oibaf~m (git-c05261a 2023-12-24 mantic-oibaf-ppa)GCC 13.2.0ext43840x2160ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorOSKernelDesktopDisplay ServerOpenGLCompilerFile-SystemScreen ResolutionDeepsparse 1.7 Raptor Lake BenchmarksSystem Logs- Transparent Huge Pages: madvise- Scaling Governor: intel_pstate powersave (EPP: performance) - CPU Microcode: 0x122 - Thermald 2.5.4- 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 + reg_file_data_sampling: Mitigation of Clear Register File + retbleed: Not affected + spec_rstack_overflow: Not affected + 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 RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected

abcdeResult OverviewPhoronix Test Suite100%102%105%107%109%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 DeepSparseR.5.S.I - S.S.SR.5.S.I - S.S.SR.5.S.I - A.M.SR.5.S.I - A.M.SL.C.7.Q - A.M.SL.C.7.Q - A.M.SN.T.C.D.m - S.S.SN.T.C.D.m - S.S.SC.S.9.P.Y.P - S.S.SC.S.9.P.Y.P - S.S.SN.D.C.o.b.u.o.I - A.M.SN.D.C.o.b.u.o.I - A.M.SC.C.R.5.I - S.S.SC.C.R.5.I - S.S.SN.D.C.o.b.u.o.I - S.S.SN.D.C.o.b.u.o.I - S.S.SC.C.R.5.I - A.M.SN.T.C.B.b.u.c - S.S.SC.C.R.5.I - A.M.SN.T.C.B.b.u.c - S.S.SN.T.C.B.b.u.S.S.I - S.S.SN.T.C.B.b.u.S.S.I - S.S.SR.5.B - S.S.SR.5.B - S.S.SC.S.9.P.Y.P - A.M.SC.S.9.P.Y.P - A.M.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.T.C.B.b.u.S.S.I - A.M.SN.T.C.B.b.u.S.S.I - A.M.SC.D.Y.C.S.I - S.S.SC.D.Y.C.S.I - S.S.SB.L.N.Q.A.S.I - A.M.SB.L.N.Q.A.S.I - A.M.SB.L.N.Q.A.S.I - S.S.SB.L.N.Q.A.S.I - S.S.SR.5.B - A.M.SR.5.B - A.M.SL.C.7.Q - S.S.SL.C.7.Q - S.S.SC.D.Y.C.S.I - A.M.SC.D.Y.C.S.I - A.M.S

deepsparse 1.7 raptor lakedeepsparse: ResNet-50, Sparse INT8 - Synchronous Single-Streamdeepsparse: ResNet-50, Sparse INT8 - Synchronous Single-Streamdeepsparse: ResNet-50, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: ResNet-50, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: Llama2 Chat 7b Quantized - Asynchronous Multi-Streamdeepsparse: Llama2 Chat 7b Quantized - 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 - Synchronous Single-Streamdeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Synchronous Single-Streamdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Streamdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Streamdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Streamdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-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 - Synchronous Single-Streamdeepsparse: ResNet-50, Baseline - Synchronous Single-Streamdeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Streamdeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-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: CV Detection, YOLOv5s COCO, Sparse INT8 - Synchronous Single-Streamdeepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - 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: ResNet-50, Baseline - Asynchronous Multi-Streamdeepsparse: ResNet-50, Baseline - Asynchronous Multi-Streamdeepsparse: Llama2 Chat 7b Quantized - Synchronous Single-Streamdeepsparse: Llama2 Chat 7b Quantized - Synchronous Single-Streamdeepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Asynchronous Multi-Streamabcde975.80771.02253.46071152.25197.6693518.245820.082549.786472.989113.6991374.657310.67738.0391124.3319104.11889.604127.9745105.8800142.91519.4444324.56303.07767.9967124.9918269.972614.815681.491249.0714379.714110.53389.1560436.248367.262414.862820.3301196.62027.9348125.919427.9492143.037888.868011.250871.697555.7746941.78631.05913.44881156.17667.7518512.689319.660750.854471.426713.9983368.585510.85167.8904126.6727104.3929.578927.7941106.2107143.83589.4148327.19123.05317.9282126.0684268.082514.919581.502149.066379.142210.54959.199434.220567.624114.783820.2407197.48387.9643125.448727.9208143.186688.870111.250671.745755.7375927.77261.07463.531129.7277.5635525.386219.646250.89271.438313.9961373.123310.71977.9405125.8707105.19349.505927.9005105.4393143.28849.4838326.8883.0567.9467125.7691267.995214.924581.508649.0612378.262610.57399.1596436.115867.381714.836920.2619197.27727.9518125.633427.8803143.391488.938111.24271.762755.72421011.71460.98563.46011152.75237.6435520.011519.822750.440572.066713.8746375.629310.64967.9734125.3512105.53789.475027.9110106.7958143.23679.3639327.29073.05217.9961125.0031269.620814.834881.682348.9564378.167610.57699.1642435.854167.411014.830520.2703197.19627.9571125.568527.9572142.998288.928011.243371.760555.7256994.76771.00243.43361161.29787.6128521.999219.958750.094271.938313.8988368.225610.86237.9248126.1129104.51839.567327.6188106.6262144.74579.3782328.56053.04047.9452125.7981268.551914.893581.174549.2619377.363910.59939.2062433.882267.319514.850420.2315197.57477.9673125.410827.8453143.570188.97811.237171.775355.7142OpenBenchmarking.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: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-Streamedcba2004006008001000SE +/- 1.14, N = 3SE +/- 10.53, N = 5994.771011.71927.77941.79975.81

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-Streamedcba0.24180.48360.72540.96721.209SE +/- 0.0010, N = 3SE +/- 0.0112, N = 51.00240.98561.07461.05911.0225

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Streamedcba0.79431.58862.38293.17723.9715SE +/- 0.0320, N = 3SE +/- 0.0108, N = 33.43363.46013.53003.44883.4607

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Streamedcba2004006008001000SE +/- 10.56, N = 3SE +/- 3.50, N = 31161.301152.751129.731156.181152.25

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: Llama2 Chat 7b Quantized - Scenario: Asynchronous Multi-Streamedcba246810SE +/- 0.0734, N = 3SE +/- 0.0698, N = 37.61287.64357.56357.75187.6693

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: Llama2 Chat 7b Quantized - Scenario: Asynchronous Multi-Streamedcba110220330440550SE +/- 4.93, N = 3SE +/- 4.73, N = 3522.00520.01525.39512.69518.25

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Streamedcba510152025SE +/- 0.09, N = 3SE +/- 0.06, N = 319.9619.8219.6519.6620.08

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Streamedcba1122334455SE +/- 0.23, N = 3SE +/- 0.14, N = 350.0950.4450.8950.8549.79

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Streamedcba1632486480SE +/- 0.34, N = 3SE +/- 0.25, N = 371.9472.0771.4471.4372.99

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Streamedcba48121620SE +/- 0.06, N = 3SE +/- 0.05, N = 313.9013.8714.0014.0013.70

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Streamedcba80160240320400SE +/- 3.00, N = 3SE +/- 3.10, N = 3368.23375.63373.12368.59374.66

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Streamedcba3691215SE +/- 0.09, N = 3SE +/- 0.09, N = 310.8610.6510.7210.8510.68

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Streamedcba246810SE +/- 0.0250, N = 3SE +/- 0.0189, N = 37.92487.97347.94057.89048.0391

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Streamedcba306090120150SE +/- 0.39, N = 3SE +/- 0.29, N = 3126.11125.35125.87126.67124.33

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Streamedcba20406080100SE +/- 0.16, N = 3SE +/- 0.22, N = 3104.52105.54105.19104.39104.12

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Streamedcba3691215SE +/- 0.0143, N = 3SE +/- 0.0203, N = 39.56739.47509.50599.57899.6041

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Streamedcba714212835SE +/- 0.04, N = 3SE +/- 0.03, N = 327.6227.9127.9027.7927.97

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Streamedcba20406080100SE +/- 0.61, N = 3SE +/- 0.24, N = 3106.63106.80105.44106.21105.88

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Streamedcba306090120150SE +/- 0.22, N = 3SE +/- 0.17, N = 3144.75143.24143.29143.84142.92

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Streamedcba3691215SE +/- 0.0538, N = 3SE +/- 0.0212, N = 39.37829.36399.48389.41489.4444

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Synchronous Single-Streamedcba70140210280350SE +/- 0.60, N = 3SE +/- 0.47, N = 3328.56327.29326.89327.19324.56

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Synchronous Single-Streamedcba0.69251.3852.07752.773.4625SE +/- 0.0056, N = 3SE +/- 0.0044, N = 33.04043.05213.05603.05313.0776

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Baseline - Scenario: Synchronous Single-Streamedcba246810SE +/- 0.0415, N = 3SE +/- 0.0195, N = 37.94527.99617.94677.92827.9967

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Baseline - Scenario: Synchronous Single-Streamedcba306090120150SE +/- 0.65, N = 3SE +/- 0.30, N = 3125.80125.00125.77126.07124.99

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Streamedcba60120180240300SE +/- 0.87, N = 3SE +/- 1.11, N = 3268.55269.62268.00268.08269.97

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Streamedcba48121620SE +/- 0.05, N = 3SE +/- 0.06, N = 314.8914.8314.9214.9214.82

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Streamedcba20406080100SE +/- 0.22, N = 3SE +/- 0.08, N = 381.1781.6881.5181.5081.49

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Streamedcba1122334455SE +/- 0.13, N = 3SE +/- 0.05, N = 349.2648.9649.0649.0749.07

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Streamedcba80160240320400SE +/- 1.11, N = 3SE +/- 1.12, N = 3377.36378.17378.26379.14379.71

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Streamedcba3691215SE +/- 0.03, N = 3SE +/- 0.03, N = 310.6010.5810.5710.5510.53

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Streamedcba3691215SE +/- 0.0105, N = 3SE +/- 0.0346, N = 39.20629.16429.15969.19909.1560

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Streamedcba90180270360450SE +/- 0.49, N = 3SE +/- 1.64, N = 3433.88435.85436.12434.22436.25

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-Streamedcba1530456075SE +/- 0.14, N = 3SE +/- 0.03, N = 367.3267.4167.3867.6267.26

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-Streamedcba48121620SE +/- 0.03, N = 3SE +/- 0.01, N = 314.8514.8314.8414.7814.86

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Streamedcba510152025SE +/- 0.02, N = 3SE +/- 0.01, N = 320.2320.2720.2620.2420.33

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Streamedcba4080120160200SE +/- 0.19, N = 3SE +/- 0.14, N = 3197.57197.20197.28197.48196.62

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Synchronous Single-Streamedcba246810SE +/- 0.0196, N = 3SE +/- 0.0106, N = 37.96737.95717.95187.96437.9348

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Synchronous Single-Streamedcba306090120150SE +/- 0.31, N = 3SE +/- 0.17, N = 3125.41125.57125.63125.45125.92

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Streamedcba714212835SE +/- 0.01, N = 3SE +/- 0.05, N = 327.8527.9627.8827.9227.95

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Streamedcba306090120150SE +/- 0.03, N = 3SE +/- 0.27, N = 3143.57143.00143.39143.19143.04

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: Llama2 Chat 7b Quantized - Scenario: Synchronous Single-Streamedcba20406080100SE +/- 0.02, N = 3SE +/- 0.03, N = 388.9888.9388.9488.8788.87

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: Llama2 Chat 7b Quantized - Scenario: Synchronous Single-Streamedcba3691215SE +/- 0.00, N = 3SE +/- 0.00, N = 311.2411.2411.2411.2511.25

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Streamedcba1632486480SE +/- 0.03, N = 3SE +/- 0.04, N = 371.7871.7671.7671.7571.70

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Streamedcba1326395265SE +/- 0.02, N = 3SE +/- 0.03, N = 355.7155.7355.7255.7455.77

44 Results Shown

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