xmas Intel Core i9-10980XE testing with a ASRock X299 Steel Legend (P1.50 BIOS) and llvmpipe on Ubuntu 22.04 via the Phoronix Test Suite. a: Processor: Intel Core i9-10980XE @ 4.80GHz (18 Cores / 36 Threads), Motherboard: ASRock X299 Steel Legend (P1.50 BIOS), Chipset: Intel Sky Lake-E DMI3 Registers, Memory: 32GB, Disk: Samsung SSD 970 PRO 512GB, Graphics: llvmpipe, Audio: Realtek ALC1220, Network: Intel I219-V + Intel I211 OS: Ubuntu 22.04, Kernel: 6.2.0-36-generic (x86_64), Desktop: GNOME Shell 42.2, Display Server: X Server 1.21.1.4, OpenGL: 4.5 Mesa 22.0.1 (LLVM 13.0.1 256 bits), Vulkan: 1.2.204, Compiler: GCC 11.4.0, File-System: ext4, Screen Resolution: 1024x768 b: Processor: Intel Core i9-10980XE @ 4.80GHz (18 Cores / 36 Threads), Motherboard: ASRock X299 Steel Legend (P1.50 BIOS), Chipset: Intel Sky Lake-E DMI3 Registers, Memory: 32GB, Disk: Samsung SSD 970 PRO 512GB, Graphics: llvmpipe, Audio: Realtek ALC1220, Network: Intel I219-V + Intel I211 OS: Ubuntu 22.04, Kernel: 6.2.0-36-generic (x86_64), Desktop: GNOME Shell 42.2, Display Server: X Server 1.21.1.4, OpenGL: 4.5 Mesa 22.0.1 (LLVM 13.0.1 256 bits), Vulkan: 1.2.204, Compiler: GCC 11.4.0, File-System: ext4, Screen Resolution: 1024x768 c: Processor: Intel Core i9-10980XE @ 4.80GHz (18 Cores / 36 Threads), Motherboard: ASRock X299 Steel Legend (P1.50 BIOS), Chipset: Intel Sky Lake-E DMI3 Registers, Memory: 32GB, Disk: Samsung SSD 970 PRO 512GB, Graphics: llvmpipe, Audio: Realtek ALC1220, Network: Intel I219-V + Intel I211 OS: Ubuntu 22.04, Kernel: 6.2.0-36-generic (x86_64), Desktop: GNOME Shell 42.2, Display Server: X Server 1.21.1.4, OpenGL: 4.5 Mesa 22.0.1 (LLVM 13.0.1 256 bits), Vulkan: 1.2.204, Compiler: GCC 11.4.0, File-System: ext4, Screen Resolution: 1024x768 LeelaChessZero 0.30 Backend: BLAS Nodes Per Second > Higher Is Better a . 99 |================================================================= b . 105 |===================================================================== c . 107 |====================================================================== LeelaChessZero 0.30 Backend: Eigen Nodes Per Second > Higher Is Better a . 82 |==================================================================== b . 86 |======================================================================= c . 80 |================================================================== Xmrig 6.21 Variant: KawPow - Hash Count: 1M H/s > Higher Is Better a . 4305.9 |=================================================================== b . 4302.8 |=================================================================== c . 4307.1 |=================================================================== Xmrig 6.21 Variant: Monero - Hash Count: 1M H/s > Higher Is Better a . 4307.8 |=================================================================== b . 4308.7 |=================================================================== c . 4302.2 |=================================================================== Xmrig 6.21 Variant: Wownero - Hash Count: 1M H/s > Higher Is Better a . 9424.3 |=================================================================== b . 9409.8 |=================================================================== c . 9344.4 |================================================================== Xmrig 6.21 Variant: GhostRider - Hash Count: 1M H/s > Higher Is Better a . 1065.3 |================================================================== b . 1047.9 |================================================================= c . 1081.0 |=================================================================== Xmrig 6.21 Variant: CryptoNight-Heavy - Hash Count: 1M H/s > Higher Is Better a . 4303.8 |=================================================================== b . 4308.7 |=================================================================== c . 4310.8 |=================================================================== Xmrig 6.21 Variant: CryptoNight-Femto UPX2 - Hash Count: 1M H/s > Higher Is Better a . 4305.2 |=================================================================== b . 4301.2 |=================================================================== c . 4299.7 |=================================================================== Java SciMark 2.2 Computational Test: Composite Mflops > Higher Is Better a . 2329.08 |=============================================================== b . 2275.42 |============================================================= c . 2451.21 |================================================================== Java SciMark 2.2 Computational Test: Monte Carlo Mflops > Higher Is Better a . 1169.97 |================================================================== b . 1169.02 |================================================================== c . 1169.33 |================================================================== Java SciMark 2.2 Computational Test: Fast Fourier Transform Mflops > Higher Is Better a . 710.66 |=================================================================== b . 684.24 |================================================================ c . 711.85 |=================================================================== Java SciMark 2.2 Computational Test: Sparse Matrix Multiply Mflops > Higher Is Better a . 1833.48 |================================================= b . 1820.44 |================================================= c . 2453.43 |================================================================== Java SciMark 2.2 Computational Test: Dense LU Matrix Factorization Mflops > Higher Is Better a . 6206.96 |================================================================== b . 5979.07 |================================================================ c . 6196.15 |================================================================== Java SciMark 2.2 Computational Test: Jacobi Successive Over-Relaxation Mflops > Higher Is Better a . 1724.33 |================================================================== b . 1724.33 |================================================================== c . 1725.30 |================================================================== WebP2 Image Encode 20220823 Encode Settings: Default MP/s > Higher Is Better a . 8.08 |=================================================================== b . 8.27 |===================================================================== c . 7.98 |=================================================================== WebP2 Image Encode 20220823 Encode Settings: Quality 75, Compression Effort 7 MP/s > Higher Is Better a . 0.19 |===================================================================== b . 0.19 |===================================================================== c . 0.19 |===================================================================== WebP2 Image Encode 20220823 Encode Settings: Quality 95, Compression Effort 7 MP/s > Higher Is Better a . 0.09 |===================================================================== b . 0.09 |===================================================================== c . 0.09 |===================================================================== WebP2 Image Encode 20220823 Encode Settings: Quality 100, Compression Effort 5 MP/s > Higher Is Better a . 5.35 |===================================================================== b . 5.28 |==================================================================== c . 5.18 |=================================================================== WebP2 Image Encode 20220823 Encode Settings: Quality 100, Lossless Compression MP/s > Higher Is Better a . 0.02 |===================================================================== b . 0.02 |===================================================================== c . 0.02 |===================================================================== Embree 4.3 Binary: Pathtracer - Model: Crown Frames Per Second > Higher Is Better a . 20.21 |==================================================================== b . 20.19 |==================================================================== c . 20.19 |==================================================================== Embree 4.3 Binary: Pathtracer ISPC - Model: Crown Frames Per Second > Higher Is Better a . 17.56 |=================================================================== b . 17.63 |==================================================================== c . 17.76 |==================================================================== Embree 4.3 Binary: Pathtracer - Model: Asian Dragon Frames Per Second > Higher Is Better a . 23.03 |==================================================================== b . 23.08 |==================================================================== c . 23.05 |==================================================================== Embree 4.3 Binary: Pathtracer - Model: Asian Dragon Obj Frames Per Second > Higher Is Better a . 20.72 |==================================================================== b . 20.73 |==================================================================== c . 20.66 |==================================================================== Embree 4.3 Binary: Pathtracer ISPC - Model: Asian Dragon Frames Per Second > Higher Is Better a . 22.86 |==================================================================== b . 22.96 |==================================================================== c . 22.90 |==================================================================== Embree 4.3 Binary: Pathtracer ISPC - Model: Asian Dragon Obj Frames Per Second > Higher Is Better a . 19.58 |==================================================================== b . 19.66 |==================================================================== c . 19.60 |==================================================================== SVT-AV1 1.8 Encoder Mode: Preset 4 - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 3.128 |==================================================================== b . 3.110 |==================================================================== c . 3.111 |==================================================================== SVT-AV1 1.8 Encoder Mode: Preset 8 - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 32.26 |==================================================================== b . 32.18 |==================================================================== c . 32.04 |==================================================================== SVT-AV1 1.8 Encoder Mode: Preset 12 - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 108.78 |================================================================== b . 109.56 |=================================================================== c . 109.85 |=================================================================== SVT-AV1 1.8 Encoder Mode: Preset 13 - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 109.94 |=================================================================== b . 108.58 |================================================================== c . 109.54 |=================================================================== SVT-AV1 1.8 Encoder Mode: Preset 4 - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 8.484 |==================================================================== b . 8.461 |==================================================================== c . 8.432 |==================================================================== SVT-AV1 1.8 Encoder Mode: Preset 8 - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 46.14 |==================================================================== b . 45.48 |=================================================================== c . 46.47 |==================================================================== SVT-AV1 1.8 Encoder Mode: Preset 12 - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 166.80 |================================================================== b . 168.78 |=================================================================== c . 165.48 |================================================================== SVT-AV1 1.8 Encoder Mode: Preset 13 - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 190.61 |================================================================== b . 193.53 |=================================================================== c . 193.61 |=================================================================== Apache Spark TPC-H 3.5 Scale Factor: 1 - Geometric Mean Of All Queries Seconds < Lower Is Better a . 1.97200676 |=============================================================== b . 1.74252536 |======================================================== c . 1.72994856 |======================================================= Apache Spark TPC-H 3.5 Scale Factor: 1 - Q01 Seconds < Lower Is Better a . 3.99205661 |=============================================================== b . 3.83936620 |============================================================= c . 3.73476624 |=========================================================== Apache Spark TPC-H 3.5 Scale Factor: 1 - Q02 Seconds < Lower Is Better a . 2.08580303 |============================================================ b . 1.88546371 |====================================================== c . 2.20615315 |=============================================================== Apache Spark TPC-H 3.5 Scale Factor: 1 - Q03 Seconds < Lower Is Better a . 2.50698948 |=============================================================== b . 2.35663772 |=========================================================== c . 2.29606605 |========================================================== Apache Spark TPC-H 3.5 Scale Factor: 1 - Q04 Seconds < Lower Is Better a . 2.41870141 |=============================================================== b . 1.97677004 |=================================================== c . 2.02678061 |===================================================== Apache Spark TPC-H 3.5 Scale Factor: 1 - Q05 Seconds < Lower Is Better a . 2.90483642 |=============================================================== b . 2.34203458 |=================================================== c . 2.50918603 |====================================================== Apache Spark TPC-H 3.5 Scale Factor: 1 - Q06 Seconds < Lower Is Better a . 0.80232513 |=============================================================== b . 0.73341227 |========================================================== c . 0.77613777 |============================================================= Apache Spark TPC-H 3.5 Scale Factor: 1 - Q07 Seconds < Lower Is Better a . 2.53732586 |=============================================================== b . 2.26433444 |======================================================== c . 2.29646873 |========================================================= Apache Spark TPC-H 3.5 Scale Factor: 1 - Q08 Seconds < Lower Is Better a . 1.94572318 |=============================================================== b . 1.87203193 |============================================================= c . 1.90735745 |============================================================== Apache Spark TPC-H 3.5 Scale Factor: 1 - Q09 Seconds < Lower Is Better a . 3.75430512 |=============================================================== b . 3.01985335 |=================================================== c . 3.19596457 |====================================================== Apache Spark TPC-H 3.5 Scale Factor: 1 - Q10 Seconds < Lower Is Better a . 2.39821315 |=============================================================== b . 1.94452596 |=================================================== c . 2.05001473 |====================================================== Apache Spark TPC-H 3.5 Scale Factor: 1 - Q11 Seconds < Lower Is Better a . 1.18394649 |=============================================================== b . 1.13731420 |============================================================= c . 0.93094862 |================================================== Apache Spark TPC-H 3.5 Scale Factor: 1 - Q12 Seconds < Lower Is Better a . 1.53979170 |=============================================================== b . 1.48578966 |============================================================= c . 1.52121258 |============================================================== Apache Spark TPC-H 3.5 Scale Factor: 1 - Q13 Seconds < Lower Is Better a . 1.63602674 |=============================================================== b . 1.23447478 |================================================ c . 1.21259868 |=============================================== Apache Spark TPC-H 3.5 Scale Factor: 1 - Q14 Seconds < Lower Is Better a . 1.20057631 |=============================================================== b . 1.09110844 |========================================================= c . 1.12718666 |=========================================================== Apache Spark TPC-H 3.5 Scale Factor: 1 - Q15 Seconds < Lower Is Better a . 1.50106275 |=============================================================== b . 1.25461686 |===================================================== c . 1.33410132 |======================================================== Apache Spark TPC-H 3.5 Scale Factor: 1 - Q16 Seconds < Lower Is Better a . 1.64840019 |=============================================================== b . 1.42126691 |====================================================== c . 1.49203193 |========================================================= Apache Spark TPC-H 3.5 Scale Factor: 1 - Q17 Seconds < Lower Is Better a . 2.26328945 |=============================================================== b . 2.11436844 |=========================================================== c . 2.15034795 |============================================================ Apache Spark TPC-H 3.5 Scale Factor: 1 - Q18 Seconds < Lower Is Better a . 3.00142813 |=============================================================== b . 2.80264425 |=========================================================== c . 2.73754430 |========================================================= Apache Spark TPC-H 3.5 Scale Factor: 1 - Q19 Seconds < Lower Is Better a . 1.15408731 |=============================================================== b . 1.06441545 |========================================================== c . 1.10046744 |============================================================ Apache Spark TPC-H 3.5 Scale Factor: 1 - Q20 Seconds < Lower Is Better a . 1.95133698 |=============================================================== b . 1.70455909 |======================================================= c . 1.72570264 |======================================================== Apache Spark TPC-H 3.5 Scale Factor: 1 - Q21 Seconds < Lower Is Better a . 7.52318430 |=============================================================== b . 6.77273893 |========================================================= c . 6.27021313 |===================================================== Apache Spark TPC-H 3.5 Scale Factor: 1 - Q22 Seconds < Lower Is Better a . 1.48474264 |=============================================================== b . 1.27632964 |====================================================== c . 1.10208726 |=============================================== Apache Spark TPC-H 3.5 Scale Factor: 10 - Geometric Mean Of All Queries Seconds < Lower Is Better a . 8.05234849 |=============================================================== b . 7.76683404 |============================================================= c . 7.81377484 |============================================================= Apache Spark TPC-H 3.5 Scale Factor: 10 - Q01 Seconds < Lower Is Better a . 9.47071648 |======================================================== b . 10.54642296 |============================================================== c . 10.40988636 |============================================================= Apache Spark TPC-H 3.5 Scale Factor: 10 - Q02 Seconds < Lower Is Better a . 5.99031353 |=============================================================== b . 4.93175220 |==================================================== c . 5.13556099 |====================================================== Apache Spark TPC-H 3.5 Scale Factor: 10 - Q03 Seconds < Lower Is Better a . 11.96 |================================================================== b . 12.32 |==================================================================== c . 12.30 |==================================================================== Apache Spark TPC-H 3.5 Scale Factor: 10 - Q04 Seconds < Lower Is Better a . 9.45135117 |============================================================ b . 9.97940636 |=============================================================== c . 9.57772446 |============================================================ Apache Spark TPC-H 3.5 Scale Factor: 10 - Q05 Seconds < Lower Is Better a . 12.44 |================================================================== b . 12.20 |================================================================= c . 12.86 |==================================================================== Apache Spark TPC-H 3.5 Scale Factor: 10 - Q06 Seconds < Lower Is Better a . 5.94231749 |=============================================================== b . 5.80461121 |============================================================== c . 5.88889503 |============================================================== Apache Spark TPC-H 3.5 Scale Factor: 10 - Q07 Seconds < Lower Is Better a . 9.72101307 |========================================================== b . 10.20696068 |============================================================= c . 10.30357361 |============================================================== Apache Spark TPC-H 3.5 Scale Factor: 10 - Q08 Seconds < Lower Is Better a . 10.65014458 |============================================================= b . 9.92212391 |========================================================= c . 10.74215603 |============================================================== Apache Spark TPC-H 3.5 Scale Factor: 10 - Q09 Seconds < Lower Is Better a . 13.57 |==================================================================== b . 13.02 |================================================================= c . 13.23 |================================================================== Apache Spark TPC-H 3.5 Scale Factor: 10 - Q10 Seconds < Lower Is Better a . 9.68672657 |=============================================================== b . 9.72717571 |=============================================================== c . 9.42293262 |============================================================= Apache Spark TPC-H 3.5 Scale Factor: 10 - Q11 Seconds < Lower Is Better a . 2.64832497 |=============================================================== b . 2.59831715 |============================================================== c . 2.48492455 |=========================================================== Apache Spark TPC-H 3.5 Scale Factor: 10 - Q12 Seconds < Lower Is Better a . 7.73613167 |=============================================================== b . 7.77267551 |=============================================================== c . 7.66917181 |============================================================== Apache Spark TPC-H 3.5 Scale Factor: 10 - Q13 Seconds < Lower Is Better a . 4.26676846 |=============================================================== b . 4.14102268 |============================================================= c . 4.26253653 |=============================================================== Apache Spark TPC-H 3.5 Scale Factor: 10 - Q14 Seconds < Lower Is Better a . 6.97334051 |=============================================================== b . 6.81118393 |============================================================== c . 6.64122009 |============================================================ Apache Spark TPC-H 3.5 Scale Factor: 10 - Q15 Seconds < Lower Is Better a . 6.55926418 |=============================================================== b . 6.53355455 |=============================================================== c . 6.55197144 |=============================================================== Apache Spark TPC-H 3.5 Scale Factor: 10 - Q16 Seconds < Lower Is Better a . 4.00392866 |=============================================================== b . 3.24821448 |=================================================== c . 3.48719382 |======================================================= Apache Spark TPC-H 3.5 Scale Factor: 10 - Q17 Seconds < Lower Is Better a . 16.38 |==================================================================== b . 16.20 |=================================================================== c . 16.19 |=================================================================== Apache Spark TPC-H 3.5 Scale Factor: 10 - Q18 Seconds < Lower Is Better a . 16.95 |==================================================================== b . 16.56 |================================================================== c . 16.66 |=================================================================== Apache Spark TPC-H 3.5 Scale Factor: 10 - Q19 Seconds < Lower Is Better a . 6.64607334 |============================================================== b . 6.55740166 |============================================================= c . 6.74200249 |=============================================================== Apache Spark TPC-H 3.5 Scale Factor: 10 - Q20 Seconds < Lower Is Better a . 8.68782139 |=============================================================== b . 8.39497089 |============================================================= c . 8.46767712 |============================================================= Apache Spark TPC-H 3.5 Scale Factor: 10 - Q21 Seconds < Lower Is Better a . 31.77 |==================================================================== b . 31.07 |=================================================================== c . 31.42 |=================================================================== Apache Spark TPC-H 3.5 Scale Factor: 10 - Q22 Seconds < Lower Is Better a . 3.39848256 |=============================================================== b . 3.23901415 |============================================================ c . 3.09575415 |========================================================= PyTorch 2.1 Device: CPU - Batch Size: 1 - Model: ResNet-50 batches/sec > Higher Is Better PyTorch 2.1 Device: CPU - Batch Size: 1 - Model: ResNet-152 batches/sec > Higher Is Better PyTorch 2.1 Device: CPU - Batch Size: 512 - Model: ResNet-50 batches/sec > Higher Is Better PyTorch 2.1 Device: CPU - Batch Size: 512 - Model: ResNet-152 batches/sec > Higher Is Better PyTorch 2.1 Device: CPU - Batch Size: 1 - Model: Efficientnet_v2_l batches/sec > Higher Is Better PyTorch 2.1 Device: CPU - Batch Size: 512 - Model: Efficientnet_v2_l batches/sec > Higher Is Better Neural Magic DeepSparse 1.6 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 18.04 |=================================================================== b . 18.44 |==================================================================== c . 18.26 |=================================================================== Neural Magic DeepSparse 1.6 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 497.07 |=================================================================== b . 485.85 |================================================================= c . 492.47 |================================================================== Neural Magic DeepSparse 1.6 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 15.91 |=================================================================== b . 16.01 |=================================================================== c . 16.25 |==================================================================== Neural Magic DeepSparse 1.6 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 62.83 |==================================================================== b . 62.45 |==================================================================== c . 61.52 |=================================================================== Neural Magic DeepSparse 1.6 Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 456.68 |=================================================================== b . 456.12 |=================================================================== c . 455.25 |=================================================================== Neural Magic DeepSparse 1.6 Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 19.68 |==================================================================== b . 19.71 |==================================================================== c . 19.75 |==================================================================== Neural Magic DeepSparse 1.6 Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 146.96 |================================================================== b . 145.39 |================================================================== c . 148.27 |=================================================================== Neural Magic DeepSparse 1.6 Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 6.7945 |================================================================== b . 6.8689 |=================================================================== c . 6.7349 |================================================================== Neural Magic DeepSparse 1.6 Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 255.58 |================================================================== b . 257.86 |=================================================================== c . 258.83 |=================================================================== Neural Magic DeepSparse 1.6 Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 35.19 |==================================================================== b . 34.88 |=================================================================== c . 34.75 |=================================================================== Neural Magic DeepSparse 1.6 Model: ResNet-50, Baseline - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 126.26 |=============================================================== b . 134.83 |=================================================================== c . 123.73 |============================================================= Neural Magic DeepSparse 1.6 Model: ResNet-50, Baseline - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 7.9078 |================================================================== b . 7.4046 |============================================================= c . 8.0707 |=================================================================== Neural Magic DeepSparse 1.6 Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 1200.72 |================================================================== b . 1203.57 |================================================================== c . 1201.85 |================================================================== Neural Magic DeepSparse 1.6 Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 7.4763 |=================================================================== b . 7.4584 |=================================================================== c . 7.4689 |=================================================================== Neural Magic DeepSparse 1.6 Model: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 566.71 |================================================================= b . 584.81 |=================================================================== c . 569.80 |================================================================= Neural Magic DeepSparse 1.6 Model: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 1.7553 |=================================================================== b . 1.7004 |================================================================= c . 1.7464 |=================================================================== Neural Magic DeepSparse 1.6 Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 104.29 |================================================================== b . 105.60 |=================================================================== c . 105.42 |=================================================================== Neural Magic DeepSparse 1.6 Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 86.27 |==================================================================== b . 85.20 |=================================================================== c . 85.34 |=================================================================== Neural Magic DeepSparse 1.6 Model: CV Detection, YOLOv5s COCO - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 83.53 |=================================================================== b . 84.21 |==================================================================== c . 82.78 |=================================================================== Neural Magic DeepSparse 1.6 Model: CV Detection, YOLOv5s COCO - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 11.96 |=================================================================== b . 11.86 |=================================================================== c . 12.06 |==================================================================== Neural Magic DeepSparse 1.6 Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 23.48 |==================================================================== b . 23.57 |==================================================================== c . 23.48 |==================================================================== Neural Magic DeepSparse 1.6 Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 383.34 |=================================================================== b . 381.84 |=================================================================== c . 383.23 |=================================================================== Neural Magic DeepSparse 1.6 Model: BERT-Large, NLP Question Answering - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 15.01 |==================================================================== b . 15.10 |==================================================================== c . 14.94 |=================================================================== Neural Magic DeepSparse 1.6 Model: BERT-Large, NLP Question Answering - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 66.63 |==================================================================== b . 66.21 |=================================================================== c . 66.91 |==================================================================== Neural Magic DeepSparse 1.6 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 260.43 |=================================================================== b . 256.18 |================================================================== c . 257.21 |================================================================== Neural Magic DeepSparse 1.6 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 34.52 |=================================================================== b . 35.11 |==================================================================== c . 34.97 |==================================================================== Neural Magic DeepSparse 1.6 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 130.11 |=================================================================== b . 121.15 |============================================================== c . 129.20 |=================================================================== Neural Magic DeepSparse 1.6 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 7.6740 |============================================================== b . 8.2423 |=================================================================== c . 7.7283 |=============================================================== Neural Magic DeepSparse 1.6 Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 112.27 |=================================================================== b . 110.23 |================================================================== c . 110.71 |================================================================== Neural Magic DeepSparse 1.6 Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 80.10 |=================================================================== b . 81.63 |==================================================================== c . 81.27 |==================================================================== Neural Magic DeepSparse 1.6 Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 84.21 |==================================================================== b . 84.27 |==================================================================== c . 84.45 |==================================================================== Neural Magic DeepSparse 1.6 Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 11.86 |==================================================================== b . 11.86 |==================================================================== c . 11.83 |==================================================================== Neural Magic DeepSparse 1.6 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 151.55 |================================================================== b . 148.28 |================================================================= c . 153.24 |=================================================================== Neural Magic DeepSparse 1.6 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 59.37 |=================================================================== b . 60.67 |==================================================================== c . 58.71 |================================================================== Neural Magic DeepSparse 1.6 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 88.75 |=================================================================== b . 87.13 |================================================================== c . 89.47 |==================================================================== Neural Magic DeepSparse 1.6 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 11.26 |=================================================================== b . 11.47 |==================================================================== c . 11.17 |================================================================== Neural Magic DeepSparse 1.6 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 22.33 |==================================================================== b . 21.89 |=================================================================== c . 22.27 |==================================================================== Neural Magic DeepSparse 1.6 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 403.05 |================================================================== b . 408.05 |=================================================================== c . 404.03 |================================================================== Neural Magic DeepSparse 1.6 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 21.89 |==================================================================== b . 21.14 |================================================================== c . 21.94 |==================================================================== Neural Magic DeepSparse 1.6 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 45.66 |================================================================== b . 47.29 |==================================================================== c . 45.55 |================================================================= Neural Magic DeepSparse 1.6 Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 209.78 |=================================================================== b . 206.48 |================================================================== c . 210.21 |=================================================================== Neural Magic DeepSparse 1.6 Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 42.88 |=================================================================== b . 43.56 |==================================================================== c . 42.79 |=================================================================== Neural Magic DeepSparse 1.6 Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 52.67 |============================================================== b . 57.40 |==================================================================== c . 56.75 |=================================================================== Neural Magic DeepSparse 1.6 Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 18.98 |==================================================================== b . 17.41 |============================================================== c . 17.61 |=============================================================== Neural Magic DeepSparse 1.6 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 18.61 |==================================================================== b . 18.18 |================================================================== c . 18.20 |=================================================================== Neural Magic DeepSparse 1.6 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 481.55 |================================================================= b . 493.56 |=================================================================== c . 493.53 |=================================================================== Neural Magic DeepSparse 1.6 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 16.40 |==================================================================== b . 16.30 |=================================================================== c . 16.48 |==================================================================== Neural Magic DeepSparse 1.6 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 60.96 |==================================================================== b . 61.36 |==================================================================== c . 60.68 |=================================================================== ScyllaDB 5.2.9 Test: Writes Op/s > Higher Is Better a . 150834 |=================================================================== b . 90383 |======================================== c . 150956 |===================================================================