Ryzen 7 7700X Tests for a future article. AMD Ryzen 7 7700X 8-Core testing with a ASRock X670E PG Lightning (1.11 BIOS) and XFX AMD Radeon RX 6400 4GB on Ubuntu 22.04 via the Phoronix Test Suite. A: Processor: AMD Ryzen 7 7700X 8-Core @ 5.57GHz (8 Cores / 16 Threads), Motherboard: ASRock X670E PG Lightning (1.11 BIOS), Chipset: AMD Device 14d8, Memory: 32GB, Disk: 1000GB Western Digital WDS100T1X0E-00AFY0, Graphics: XFX AMD Radeon RX 6400 4GB (2320/1000MHz), Audio: AMD Navi 21 HDMI Audio, Monitor: ASUS MG28U, Network: Realtek RTL8125 2.5GbE OS: Ubuntu 22.04, Kernel: 5.17.0-1013-oem (x86_64), Desktop: GNOME Shell 42.2, Display Server: X Server 1.21.1.3 + Wayland, OpenGL: 4.6 Mesa 22.2.0-devel (git-a9610ab740) (LLVM 13.0.1 DRM 3.44), Vulkan: 1.3.219, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 3840x2160 B: Processor: AMD Ryzen 7 7700X 8-Core @ 5.57GHz (8 Cores / 16 Threads), Motherboard: ASRock X670E PG Lightning (1.11 BIOS), Chipset: AMD Device 14d8, Memory: 32GB, Disk: 1000GB Western Digital WDS100T1X0E-00AFY0, Graphics: XFX AMD Radeon RX 6400 4GB (2320/1000MHz), Audio: AMD Navi 21 HDMI Audio, Monitor: ASUS MG28U, Network: Realtek RTL8125 2.5GbE OS: Ubuntu 22.04, Kernel: 5.17.0-1013-oem (x86_64), Desktop: GNOME Shell 42.2, Display Server: X Server 1.21.1.3 + Wayland, OpenGL: 4.6 Mesa 22.2.0-devel (git-a9610ab740) (LLVM 13.0.1 DRM 3.44), Vulkan: 1.3.219, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 3840x2160 SMHasher 2022-08-22 Hash: wyhash MiB/sec > Higher Is Better A . 24952.00 |================================================================= B . 24812.67 |================================================================= SMHasher 2022-08-22 Hash: wyhash cycles/hash < Lower Is Better A . 18.00 |=================================================================== B . 18.23 |==================================================================== SMHasher 2022-08-22 Hash: SHA3-256 MiB/sec > Higher Is Better A . 171.75 |=================================================================== B . 170.07 |================================================================== SMHasher 2022-08-22 Hash: SHA3-256 cycles/hash < Lower Is Better A . 2314.14 |================================================================= B . 2340.81 |================================================================== SMHasher 2022-08-22 Hash: Spooky32 MiB/sec > Higher Is Better A . 16441.99 |================================================================= B . 16367.89 |================================================================= SMHasher 2022-08-22 Hash: Spooky32 cycles/hash < Lower Is Better A . 33.91 |==================================================================== B . 34.11 |==================================================================== SMHasher 2022-08-22 Hash: fasthash32 MiB/sec > Higher Is Better A . 6990.21 |================================================================== B . 6957.30 |================================================================== SMHasher 2022-08-22 Hash: fasthash32 cycles/hash < Lower Is Better A . 27.77 |==================================================================== B . 27.92 |==================================================================== SMHasher 2022-08-22 Hash: FarmHash128 MiB/sec > Higher Is Better A . 15031.76 |================================================================= B . 14949.14 |================================================================= SMHasher 2022-08-22 Hash: FarmHash128 cycles/hash < Lower Is Better A . 58.62 |==================================================================== B . 58.98 |==================================================================== SMHasher 2022-08-22 Hash: t1ha2_atonce MiB/sec > Higher Is Better A . 16567.35 |================================================================= B . 16469.72 |================================================================= SMHasher 2022-08-22 Hash: t1ha2_atonce cycles/hash < Lower Is Better A . 25.84 |==================================================================== B . 25.85 |==================================================================== SMHasher 2022-08-22 Hash: FarmHash32 x86_64 AVX MiB/sec > Higher Is Better A . 29625.18 |================================================================= B . 29456.09 |================================================================= SMHasher 2022-08-22 Hash: FarmHash32 x86_64 AVX cycles/hash < Lower Is Better A . 33.00 |==================================================================== B . 33.10 |==================================================================== SMHasher 2022-08-22 Hash: t1ha0_aes_avx2 x86_64 MiB/sec > Higher Is Better A . 81587.16 |================================================================= B . 81123.62 |================================================================= SMHasher 2022-08-22 Hash: t1ha0_aes_avx2 x86_64 cycles/hash < Lower Is Better A . 25.55 |=================================================================== B . 25.77 |==================================================================== SMHasher 2022-08-22 Hash: MeowHash x86_64 AES-NI MiB/sec > Higher Is Better A . 43409.48 |================================================================= B . 43176.20 |================================================================= SMHasher 2022-08-22 Hash: MeowHash x86_64 AES-NI cycles/hash < Lower Is Better A . 56.05 |==================================================================== B . 56.35 |==================================================================== OpenRadioss 2022.10.13 Model: Bumper Beam Seconds < Lower Is Better A . 120.80 |=========================================================== B . 136.59 |=================================================================== OpenRadioss 2022.10.13 Model: Cell Phone Drop Test Seconds < Lower Is Better A . 82.66 |============================================================ B . 93.67 |==================================================================== OpenRadioss 2022.10.13 Model: Bird Strike on Windshield Seconds < Lower Is Better A . 252.55 |============================================================= B . 277.84 |=================================================================== OpenRadioss 2022.10.13 Model: Rubber O-Ring Seal Installation Seconds < Lower Is Better A . 119.08 |======================================================== B . 143.60 |=================================================================== OpenRadioss 2022.10.13 Model: INIVOL and Fluid Structure Interaction Drop Container Seconds < Lower Is Better A . 526.03 |============================================================== B . 565.63 |=================================================================== Xmrig 6.18.1 Variant: Monero - Hash Count: 1M H/s > Higher Is Better A . 7555.0 |=================================================================== B . 7050.3 |=============================================================== Xmrig 6.18.1 Variant: Wownero - Hash Count: 1M H/s > Higher Is Better A . 9679.3 |=================================================================== B . 9558.7 |================================================================== JPEG XL libjxl 0.7 Input: PNG - Quality: 80 MP/s > Higher Is Better A . 11.44 |==================================================================== B . 11.37 |==================================================================== JPEG XL libjxl 0.7 Input: PNG - Quality: 90 MP/s > Higher Is Better A . 11.27 |==================================================================== B . 11.19 |==================================================================== JPEG XL libjxl 0.7 Input: JPEG - Quality: 80 MP/s > Higher Is Better A . 11.12 |==================================================================== B . 11.03 |=================================================================== JPEG XL libjxl 0.7 Input: JPEG - Quality: 90 MP/s > Higher Is Better A . 10.96 |==================================================================== B . 10.90 |==================================================================== JPEG XL libjxl 0.7 Input: PNG - Quality: 100 MP/s > Higher Is Better A . 0.90 |==================================================================== B . 0.91 |===================================================================== JPEG XL libjxl 0.7 Input: JPEG - Quality: 100 MP/s > Higher Is Better A . 0.78 |================================================================== B . 0.82 |===================================================================== JPEG XL Decoding libjxl 0.7 CPU Threads: 1 MP/s > Higher Is Better A . 64.73 |=================================================================== B . 65.81 |==================================================================== JPEG XL Decoding libjxl 0.7 CPU Threads: All MP/s > Higher Is Better A . 293.80 |============================================================== B . 318.09 |=================================================================== QuadRay 2022.05.25 Scene: 1 - Resolution: 4K FPS > Higher Is Better A . 12.96 |==================================================================== B . 13.02 |==================================================================== QuadRay 2022.05.25 Scene: 2 - Resolution: 4K FPS > Higher Is Better A . 3.70 |===================================================================== B . 3.69 |===================================================================== QuadRay 2022.05.25 Scene: 3 - Resolution: 4K FPS > Higher Is Better A . 3.15 |===================================================================== B . 3.14 |===================================================================== QuadRay 2022.05.25 Scene: 5 - Resolution: 4K FPS > Higher Is Better A . 0.96 |===================================================================== B . 0.89 |================================================================ QuadRay 2022.05.25 Scene: 1 - Resolution: 1080p FPS > Higher Is Better A . 49.15 |==================================================================== B . 49.44 |==================================================================== QuadRay 2022.05.25 Scene: 2 - Resolution: 1080p FPS > Higher Is Better A . 13.88 |==================================================================== B . 13.94 |==================================================================== QuadRay 2022.05.25 Scene: 3 - Resolution: 1080p FPS > Higher Is Better A . 11.69 |================================================================== B . 12.00 |==================================================================== QuadRay 2022.05.25 Scene: 5 - Resolution: 1080p FPS > Higher Is Better A . 3.90 |===================================================================== B . 3.62 |================================================================ AOM AV1 3.5 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4K Frames Per Second > Higher Is Better A . 0.26 |===================================================================== B . 0.26 |===================================================================== AOM AV1 3.5 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4K Frames Per Second > Higher Is Better A . 8.70 |===================================================================== B . 8.69 |===================================================================== AOM AV1 3.5 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K Frames Per Second > Higher Is Better A . 36.10 |==================================================================== B . 36.23 |==================================================================== AOM AV1 3.5 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K Frames Per Second > Higher Is Better A . 15.83 |==================================================================== B . 15.83 |==================================================================== AOM AV1 3.5 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K Frames Per Second > Higher Is Better A . 50.44 |=================================================================== B . 51.13 |==================================================================== AOM AV1 3.5 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K Frames Per Second > Higher Is Better A . 68.68 |=================================================================== B . 69.48 |==================================================================== AOM AV1 3.5 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K Frames Per Second > Higher Is Better A . 70.93 |==================================================================== B . 71.00 |==================================================================== AOM AV1 3.5 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080p Frames Per Second > Higher Is Better A . 0.76 |===================================================================== B . 0.75 |==================================================================== AOM AV1 3.5 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 1080p Frames Per Second > Higher Is Better A . 18.20 |==================================================================== B . 18.25 |==================================================================== AOM AV1 3.5 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080p Frames Per Second > Higher Is Better A . 71.08 |==================================================================== B . 64.53 |============================================================== AOM AV1 3.5 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080p Frames Per Second > Higher Is Better A . 49.91 |==================================================================== B . 49.74 |==================================================================== AOM AV1 3.5 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080p Frames Per Second > Higher Is Better A . 133.52 |================================================================ B . 139.06 |=================================================================== AOM AV1 3.5 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080p Frames Per Second > Higher Is Better A . 173.63 |================================================================== B . 176.02 |=================================================================== AOM AV1 3.5 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080p Frames Per Second > Higher Is Better A . 202.0 |==================================================================== B . 183.6 |============================================================== libavif avifenc 0.11 Encoder Speed: 0 Seconds < Lower Is Better A . 113.39 |=================================================================== B . 113.20 |=================================================================== libavif avifenc 0.11 Encoder Speed: 2 Seconds < Lower Is Better A . 53.78 |==================================================================== B . 53.89 |==================================================================== libavif avifenc 0.11 Encoder Speed: 6 Seconds < Lower Is Better A . 5.333 |==================================================================== B . 5.291 |=================================================================== libavif avifenc 0.11 Encoder Speed: 6, Lossless Seconds < Lower Is Better A . 8.133 |==================================================================== B . 8.018 |=================================================================== libavif avifenc 0.11 Encoder Speed: 10, Lossless Seconds < Lower Is Better A . 4.016 |================================================================== B . 4.141 |==================================================================== Y-Cruncher 0.7.10.9513 Pi Digits To Calculate: 1B Seconds < Lower Is Better A . 29.20 |==================================================================== B . 29.12 |==================================================================== Y-Cruncher 0.7.10.9513 Pi Digits To Calculate: 500M Seconds < Lower Is Better A . 13.27 |==================================================================== B . 13.17 |=================================================================== oneDNN 2.7 Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU ms < Lower Is Better A . 3.68436 |================================================================== B . 3.70584 |================================================================== oneDNN 2.7 Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU ms < Lower Is Better A . 5.36365 |================================================================== B . 5.36040 |================================================================== oneDNN 2.7 Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better A . 0.702154 |================================================================ B . 0.708957 |================================================================= oneDNN 2.7 Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better A . 1.18542 |================================================================= B . 1.20283 |================================================================== oneDNN 2.7 Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better A . 1.46205 |================================================================== B . 1.46602 |================================================================== oneDNN 2.7 Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better A . 3.03247 |================================================================== B . 3.02950 |================================================================== oneDNN 2.7 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU ms < Lower Is Better A . 9.47087 |================================================================== B . 9.45878 |================================================================== oneDNN 2.7 Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU ms < Lower Is Better A . 5.85680 |=============================================================== B . 6.10103 |================================================================== oneDNN 2.7 Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU ms < Lower Is Better A . 4.50304 |================================================================== B . 4.49511 |================================================================== oneDNN 2.7 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better A . 8.96869 |================================================================= B . 9.13721 |================================================================== oneDNN 2.7 Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better A . 0.942884 |================================================================= B . 0.943897 |================================================================= oneDNN 2.7 Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better A . 1.13062 |================================================================== B . 1.13382 |================================================================== oneDNN 2.7 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU ms < Lower Is Better A . 2618.14 |================================================================== B . 2618.26 |================================================================== oneDNN 2.7 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU ms < Lower Is Better A . 1341.09 |================================================================== B . 1346.17 |================================================================== oneDNN 2.7 Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better A . 2622.21 |================================================================== B . 2618.84 |================================================================== oneDNN 2.7 Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better A . 4.25152 |================================================================== B . 4.24314 |================================================================== oneDNN 2.7 Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better A . 9.39398 |================================================================== B . 9.32817 |================================================================== oneDNN 2.7 Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better A . 2.87135 |================================================================== B . 2.88067 |================================================================== oneDNN 2.7 Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better A . 1352.66 |================================================================== B . 1339.00 |================================================================= oneDNN 2.7 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU ms < Lower Is Better A . 1.62133 |================================================================== B . 1.63323 |================================================================== oneDNN 2.7 Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better A . 2613.39 |================================================================== B . 2618.97 |================================================================== oneDNN 2.7 Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better A . 1342.13 |================================================================== B . 1339.39 |================================================================== oneDNN 2.7 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better A . 0.315521 |================================================================ B . 0.318917 |================================================================= oneDNN 2.7 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better A . 0.532325 |================================================================= B . 0.531154 |================================================================= FLAC Audio Encoding 1.4 WAV To FLAC Seconds < Lower Is Better A . 11.73 |==================================================================== B . 11.65 |==================================================================== Cpuminer-Opt 3.20.3 Algorithm: Magi kH/s > Higher Is Better A . 536.39 |=================================================================== B . 537.52 |=================================================================== Cpuminer-Opt 3.20.3 Algorithm: x25x kH/s > Higher Is Better A . 570.19 |=================================================================== B . 563.72 |================================================================== Cpuminer-Opt 3.20.3 Algorithm: scrypt kH/s > Higher Is Better A . 315.65 |================================================================== B . 318.16 |=================================================================== Cpuminer-Opt 3.20.3 Algorithm: Deepcoin kH/s > Higher Is Better A . 10480 |==================================================================== B . 10420 |==================================================================== Cpuminer-Opt 3.20.3 Algorithm: Ringcoin kH/s > Higher Is Better A . 2220.83 |================================================================= B . 2252.00 |================================================================== Cpuminer-Opt 3.20.3 Algorithm: Blake-2 S kH/s > Higher Is Better A . 843890 |================================================================ B . 889870 |=================================================================== Cpuminer-Opt 3.20.3 Algorithm: Garlicoin kH/s > Higher Is Better A . 4639.52 |================================================================== B . 4617.41 |================================================================== Cpuminer-Opt 3.20.3 Algorithm: Skeincoin kH/s > Higher Is Better A . 134680 |=================================================================== B . 134190 |=================================================================== Cpuminer-Opt 3.20.3 Algorithm: Myriad-Groestl kH/s > Higher Is Better A . 42500 |==================================================================== B . 42210 |==================================================================== Cpuminer-Opt 3.20.3 Algorithm: LBC, LBRY Credits kH/s > Higher Is Better A . 72970 |================================================================== B . 75300 |==================================================================== Cpuminer-Opt 3.20.3 Algorithm: Quad SHA-256, Pyrite kH/s > Higher Is Better A . 162810 |=================================================================== B . 159450 |================================================================== Cpuminer-Opt 3.20.3 Algorithm: Triple SHA-256, Onecoin kH/s > Higher Is Better A . 235830 |=================================================================== B . 235080 |=================================================================== TensorFlow 2.10 Device: CPU - Batch Size: 16 - Model: AlexNet images/sec > Higher Is Better A . 121.96 |=================================================================== B . 122.86 |=================================================================== TensorFlow 2.10 Device: CPU - Batch Size: 32 - Model: AlexNet images/sec > Higher Is Better A . 165.42 |=================================================================== B . 166.20 |=================================================================== TensorFlow 2.10 Device: CPU - Batch Size: 64 - Model: AlexNet images/sec > Higher Is Better A . 199.61 |=================================================================== B . 200.32 |=================================================================== TensorFlow 2.10 Device: CPU - Batch Size: 256 - Model: AlexNet images/sec > Higher Is Better A . 229.53 |=================================================================== B . 230.41 |=================================================================== TensorFlow 2.10 Device: CPU - Batch Size: 512 - Model: AlexNet images/sec > Higher Is Better A . 231.51 |=================================================================== B . 231.79 |=================================================================== TensorFlow 2.10 Device: CPU - Batch Size: 16 - Model: GoogLeNet images/sec > Higher Is Better A . 90.58 |==================================================================== B . 90.62 |==================================================================== TensorFlow 2.10 Device: CPU - Batch Size: 16 - Model: ResNet-50 images/sec > Higher Is Better A . 29.65 |==================================================================== B . 29.67 |==================================================================== TensorFlow 2.10 Device: CPU - Batch Size: 32 - Model: GoogLeNet images/sec > Higher Is Better A . 89.76 |==================================================================== B . 90.04 |==================================================================== TensorFlow 2.10 Device: CPU - Batch Size: 32 - Model: ResNet-50 images/sec > Higher Is Better A . 29.81 |==================================================================== B . 29.76 |==================================================================== TensorFlow 2.10 Device: CPU - Batch Size: 64 - Model: GoogLeNet images/sec > Higher Is Better A . 89.21 |==================================================================== B . 89.32 |==================================================================== TensorFlow 2.10 Device: CPU - Batch Size: 64 - Model: ResNet-50 images/sec > Higher Is Better A . 29.61 |==================================================================== B . 29.67 |==================================================================== TensorFlow 2.10 Device: CPU - Batch Size: 256 - Model: GoogLeNet images/sec > Higher Is Better A . 88.97 |==================================================================== B . 88.89 |==================================================================== TensorFlow 2.10 Device: CPU - Batch Size: 256 - Model: ResNet-50 images/sec > Higher Is Better A . 29.65 |==================================================================== B . 29.61 |==================================================================== TensorFlow 2.10 Device: CPU - Batch Size: 512 - Model: GoogLeNet images/sec > Higher Is Better A . 88.99 |==================================================================== B . 89.08 |==================================================================== TensorFlow 2.10 Device: CPU - Batch Size: 512 - Model: ResNet-50 images/sec > Higher Is Better Neural Magic DeepSparse 1.1 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better A . 6.1328 |=================================================================== B . 6.1783 |=================================================================== Neural Magic DeepSparse 1.1 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better A . 649.69 |=================================================================== B . 645.72 |=================================================================== Neural Magic DeepSparse 1.1 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream items/sec > Higher Is Better A . 6.0820 |=================================================================== B . 6.0845 |=================================================================== Neural Magic DeepSparse 1.1 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better A . 164.41 |=================================================================== B . 164.34 |=================================================================== Neural Magic DeepSparse 1.1 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better A . 53.73 |==================================================================== B . 53.60 |==================================================================== Neural Magic DeepSparse 1.1 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better A . 74.42 |==================================================================== B . 74.58 |==================================================================== Neural Magic DeepSparse 1.1 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream items/sec > Higher Is Better A . 54.85 |==================================================================== B . 54.60 |==================================================================== Neural Magic DeepSparse 1.1 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better A . 18.22 |==================================================================== B . 18.31 |==================================================================== Neural Magic DeepSparse 1.1 Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better A . 63.07 |==================================================================== B . 61.62 |================================================================== Neural Magic DeepSparse 1.1 Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better A . 63.35 |================================================================== B . 64.85 |==================================================================== Neural Magic DeepSparse 1.1 Model: CV Detection,YOLOv5s COCO - Scenario: Synchronous Single-Stream items/sec > Higher Is Better A . 57.13 |==================================================================== B . 57.04 |==================================================================== Neural Magic DeepSparse 1.1 Model: CV Detection,YOLOv5s COCO - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better A . 17.49 |==================================================================== B . 17.52 |==================================================================== Neural Magic DeepSparse 1.1 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better A . 143.58 |=================================================================== B . 143.25 |=================================================================== Neural Magic DeepSparse 1.1 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better A . 27.84 |==================================================================== B . 27.90 |==================================================================== Neural Magic DeepSparse 1.1 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream items/sec > Higher Is Better A . 117.13 |=================================================================== B . 116.82 |=================================================================== Neural Magic DeepSparse 1.1 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better A . 8.5304 |=================================================================== B . 8.5531 |=================================================================== Neural Magic DeepSparse 1.1 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better A . 92.19 |==================================================================== B . 92.17 |==================================================================== Neural Magic DeepSparse 1.1 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better A . 43.38 |==================================================================== B . 43.38 |==================================================================== Neural Magic DeepSparse 1.1 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream items/sec > Higher Is Better A . 76.29 |==================================================================== B . 76.30 |==================================================================== Neural Magic DeepSparse 1.1 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better A . 13.10 |==================================================================== B . 13.10 |==================================================================== Neural Magic DeepSparse 1.1 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better A . 46.78 |==================================================================== B . 47.10 |==================================================================== Neural Magic DeepSparse 1.1 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better A . 85.48 |==================================================================== B . 84.90 |==================================================================== Neural Magic DeepSparse 1.1 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream items/sec > Higher Is Better A . 38.97 |==================================================================== B . 38.81 |==================================================================== Neural Magic DeepSparse 1.1 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better A . 25.65 |==================================================================== B . 25.76 |==================================================================== Neural Magic DeepSparse 1.1 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better A . 6.1219 |=================================================================== B . 6.1062 |=================================================================== Neural Magic DeepSparse 1.1 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better A . 648.55 |=================================================================== B . 649.40 |=================================================================== Neural Magic DeepSparse 1.1 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream items/sec > Higher Is Better A . 6.1015 |=================================================================== B . 6.1128 |=================================================================== Neural Magic DeepSparse 1.1 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better A . 163.89 |=================================================================== B . 163.58 |=================================================================== spaCy 3.4.1 Model: en_core_web_lg tokens/sec > Higher Is Better A . 18697 |==================================================================== B . 18445 |=================================================================== spaCy 3.4.1 Model: en_core_web_trf tokens/sec > Higher Is Better A . 969 |====================================================================== B . 975 |======================================================================