nvidia rtx 5090 compute benchmarks Tests for a future article. Intel Core Ultra 9 285K testing with a ASUS ROG MAXIMUS Z890 HERO (1203 BIOS) and ASUS NVIDIA GeForce RTX 5090 32GB on Ubuntu 24.10 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2501242-PTS-NVIDIART00&grs&sor .
nvidia rtx 5090 compute benchmarks Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server Display Driver OpenGL OpenCL Compiler File-System Screen Resolution rtx 5090 NVIDIA 5090 GeForce RTX 5090 Intel Core Ultra 9 285K @ 5.10GHz (24 Cores) ASUS ROG MAXIMUS Z890 HERO (1203 BIOS) Intel Device ae7f 2 x 16GB DDR5-6400MT/s Micron CP16G64C38U5B.M8D1 1000GB Western Digital WDS100T1X0E-00AFY0 + 4001GB Western Digital WD_BLACK SN850X 4000GB ASUS NVIDIA GeForce RTX 5090 32GB Intel Device 7f50 ASUS VP28U Realtek Device 8126 + Intel I226-V + Intel Wi-Fi 7 Ubuntu 24.10 6.11.0-13-generic (x86_64) GNOME Shell 47.0 X Server 1.21.1.13 NVIDIA 570.86.10 4.6.0 OpenCL 3.0 CUDA 12.8.51 + OpenCL 3.0 GCC 14.2.0 ext4 3840x2160 OpenBenchmarking.org Kernel Details - nouveau.modeset=0 - Transparent Huge Pages: madvise Compiler Details - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2,rust --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-14-zdkDXv/gcc-14-14.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-14-zdkDXv/gcc-14-14.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Processor Details - Scaling Governor: intel_pstate performance (EPP: default) - CPU Microcode: 0x114 - Thermald 2.5.8 Graphics Details - BAR1 / Visible vRAM Size: 32768 MiB - vBIOS Version: 98.02.2e.00.03 OpenCL Details - GPU Compute Cores: 21760 Security Details - 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: Not affected + 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: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected
nvidia rtx 5090 compute benchmarks ncnn: Vulkan GPU - mnasnet ncnn: Vulkan GPU - regnety_400m ncnn: Vulkan GPU - googlenet ncnn: Vulkan GPU - efficientnet-b0 ncnn: Vulkan GPU-v2-v2 - mobilenet-v2 ncnn: Vulkan GPU - FastestDet ncnn: Vulkan GPU - shufflenet-v2 ncnn: Vulkan GPU - squeezenet_ssd ncnn: Vulkan GPU - resnet18 ncnn: Vulkan GPU - blazeface ncnn: Vulkan GPU-v3-v3 - mobilenet-v3 ncnn: Vulkan GPUv2-yolov3v2-yolov3 - mobilenetv2-yolov3 ncnn: Vulkan GPU - mobilenet ncnn: Vulkan GPU - yolov4-tiny ncnn: Vulkan GPU - resnet50 realsr-ncnn: 4x - No ncnn: Vulkan GPU - alexnet ncnn: Vulkan GPU - vgg16 namd-cuda: ATPase Simulation - 327,506 Atoms blender: BMW27 - NVIDIA OptiX shoc: OpenCL - Bus Speed Readback vkfft: FFT + iFFT C2C 1D batched in half precision vkfft: FFT + iFFT C2C 1D batched in single precision vkfft: FFT + iFFT C2C 1D batched in double precision blender: BMW27 - NVIDIA CUDA waifu2x-ncnn: 2x - 3 - Yes blender: Barbershop - NVIDIA OptiX blender: Pabellon Barcelona - NVIDIA OptiX vkfft: FFT + iFFT C2C 1D batched in single precision, no reshuffling clpeak: Transfer Bandwidth enqueueReadBuffer v-ray: NVIDIA CUDA GPU hashcat: MD5 blender: Pabellon Barcelona - NVIDIA CUDA shoc: OpenCL - FFT SP blender: Barbershop - NVIDIA CUDA clpeak: Transfer Bandwidth enqueueWriteBuffer blender: Classroom - NVIDIA OptiX opencl-benchmark: Memory Bandwidth Coalesced Read opencl-benchmark: Memory Bandwidth Coalesced Write indigobench: OpenCL GPU - Bedroom hashcat: 7-Zip hashcat: SHA1 shoc: OpenCL - Triad shoc: OpenCL - S3D vkfft: FFT + iFFT C2C Bluestein in single precision hashcat: TrueCrypt RIPEMD160 + XTS blender: Junkshop - NVIDIA CUDA shoc: OpenCL - GEMM SGEMM_N blender: Fishy Cat - NVIDIA OptiX indigobench: OpenCL GPU - Supercar shoc: OpenCL - MD5 Hash clpeak: Kernel Latency opencl-benchmark: INT64 Compute blender: Junkshop - NVIDIA OptiX opencl-benchmark: INT8 Compute shoc: OpenCL - Texture Read Bandwidth ncnn: Vulkan GPU - vision_transformer opencl-benchmark: INT16 Compute realsr-ncnn: 4x - Yes clpeak: Global Memory Bandwidth fluidx3d: FP32-FP16C clpeak: Integer 24-bit Compute clpeak: Integer Compute vkfft: FFT + iFFT C2C Bluestein benchmark in double precision vkfft: FFT + iFFT R2C / C2R shoc: OpenCL - Max SP Flops vkfft: FFT + iFFT C2C multidimensional in single precision hashcat: SHA-512 opencl-benchmark: FP64 Compute vkresample: 2x - Double vkpeak: fp32-vec4 vkpeak: int32-vec4 vkpeak: int16-scalar vkpeak: fp32-scalar fluidx3d: FP32-FP32 shoc: OpenCL - Reduction opencl-benchmark: FP32 Compute opencl-benchmark: FP16 Compute clpeak: Double-Precision Compute vkpeak: fp16-vec4 vkpeak: fp16-scalar opencl-benchmark: INT32 Compute fluidx3d: FP32-FP16S vkpeak: fp64-vec4 clpeak: Single-Precision Compute vkresample: 2x - Single vkpeak: int16-vec4 vkpeak: fp64-scalar shoc: OpenCL - Bus Speed Download vkpeak: int32-scalar v-ray: NVIDIA RTX GPU blender: Fishy Cat - NVIDIA CUDA blender: Classroom - NVIDIA CUDA waifu2x-ncnn: 2x - 3 - No rtx 5090 NVIDIA 5090 GeForce RTX 5090 9.8 47.9 13.92 27.44 10.21 11.01 4.53 22.02 10.93 2.67 4.68 42.44 42.44 39.34 28.58 4.63 8.74 37.05 0.05810 2.92 28.6895 302221 237717 63738 4.72 2.299 24.33 7 243937 13.83 4851 106848250000 17.35 4398.39 35.14 18.41 6.16 1596.24 1687.49 42.759 3272300 68852500000 27.8329 1117.54 36054 2776000 8.99 35937.2 4.55 92.7 142.407 5.15 4.396 5.66 41.795 2870.68 62.76 54.018 13.484 1562.97 19140 61843.11 62151.94 9931 164933 124615 144624 8900400000 1.95 103.505 83296.84 61885.37 40006.6 63013.32 9524 837.207 117.847 122.914 1976.9 72592.93 62611.78 61.759 18499 1965.7 121415.53 5.648 43806.13 1967.37 28.7867 62142.56 11923 8.92 8.38 4.77 52.59 12.78 20.78 11.25 8.68 4.04 26.39 9.41 3.05 4.28 39.74 39.74 38.91 30.1 4.4 8.55 38.73 0.05851 2.97 28.5856 305671 235884 62773 4.72 2.267 24.6 7.06 243913 13.89 4882 106216550000 17.34 4400.85 35.28 18.42 6.14 1603.93 1679.44 42.852 3276600 69072700000 27.8098 1120.51 36026 2770700 8.98 36016.5 4.56 92.52 142.24 5.16 4.4 5.67 41.724 2872.53 62.86 54.037 13.504 1564.49 19121 61866.86 62119.51 9940 164801 124556 144600 8901200000 1.95 103.478 83290.22 61894.95 39989.7 63035.62 9527 836.98 117.864 122.944 1977.26 72578.62 62611.17 61.759 18496 1965.39 121438.41 5.649 43799.96 1967.49 28.787 62142.07 11923 8.92 8.38 6.55 36.2 18.06 19.79 8.37 9.29 5.12 25.57 11.01 2.67 4.29 38.91 38.91 36.86 29.47 4.446 8.99 37.49 0.05943 2.98 28.1406 300737 239575 63637 4.65 2.278 24.5 7.04 241913 13.78 4882 106544000000 17.44 4375.95 35.1 18.33 6.17 1596.88 1680.23 42.924 3264300 69104300000 27.7455 1120.88 35954 2777600 9 35961.3 4.55 92.711 142.532 5.15 4.392 5.66 41.757 2875.5 62.83 53.953 13.486 1564.61 19135 61903.93 62178.05 9934 164873 124646 144695 8895900000 1.951 103.453 83257.59 61914.51 39998.42 63035.62 9525 837.243 117.881 122.941 1976.78 72575.33 62597.51 61.773 18500 1965.32 121419.57 5.648 43803.6 1967.43 28.7881 62141.02 11923 8.92 8.38 OpenBenchmarking.org
NCNN Target: Vulkan GPU - Model: mnasnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20241226 Target: Vulkan GPU - Model: mnasnet NVIDIA 5090 GeForce RTX 5090 rtx 5090 3 6 9 12 15 4.77 6.55 9.80 MIN: 3.69 / MAX: 54.04 MIN: 3.69 / MAX: 59.21 MIN: 3.75 / MAX: 63.5 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: regnety_400m OpenBenchmarking.org ms, Fewer Is Better NCNN 20241226 Target: Vulkan GPU - Model: regnety_400m GeForce RTX 5090 rtx 5090 NVIDIA 5090 12 24 36 48 60 36.20 47.90 52.59 MIN: 21.98 / MAX: 458.49 MIN: 21.96 / MAX: 421.33 MIN: 21.91 / MAX: 425.58 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: googlenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20241226 Target: Vulkan GPU - Model: googlenet NVIDIA 5090 rtx 5090 GeForce RTX 5090 4 8 12 16 20 12.78 13.92 18.06 MIN: 7.62 / MAX: 95.87 MIN: 7.49 / MAX: 98.37 MIN: 7.48 / MAX: 98.98 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: efficientnet-b0 OpenBenchmarking.org ms, Fewer Is Better NCNN 20241226 Target: Vulkan GPU - Model: efficientnet-b0 GeForce RTX 5090 NVIDIA 5090 rtx 5090 6 12 18 24 30 19.79 20.78 27.44 MIN: 6.3 / MAX: 110.05 MIN: 6.36 / MAX: 109.98 MIN: 6.34 / MAX: 109.98 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20241226 Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2 GeForce RTX 5090 rtx 5090 NVIDIA 5090 3 6 9 12 15 8.37 10.21 11.25 MIN: 3.84 / MAX: 63.9 MIN: 3.85 / MAX: 64.73 MIN: 3.82 / MAX: 63.94 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: FastestDet OpenBenchmarking.org ms, Fewer Is Better NCNN 20241226 Target: Vulkan GPU - Model: FastestDet NVIDIA 5090 GeForce RTX 5090 rtx 5090 3 6 9 12 15 8.68 9.29 11.01 MIN: 5.07 / MAX: 85.5 MIN: 5.01 / MAX: 89.47 MIN: 5.09 / MAX: 92.36 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: shufflenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20241226 Target: Vulkan GPU - Model: shufflenet-v2 NVIDIA 5090 rtx 5090 GeForce RTX 5090 1.152 2.304 3.456 4.608 5.76 4.04 4.53 5.12 MIN: 3.9 / MAX: 5.77 MIN: 3.88 / MAX: 57.59 MIN: 3.91 / MAX: 67.5 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: squeezenet_ssd OpenBenchmarking.org ms, Fewer Is Better NCNN 20241226 Target: Vulkan GPU - Model: squeezenet_ssd rtx 5090 GeForce RTX 5090 NVIDIA 5090 6 12 18 24 30 22.02 25.57 26.39 MIN: 7.41 / MAX: 92.66 MIN: 7.22 / MAX: 94.51 MIN: 7.39 / MAX: 95.48 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: resnet18 OpenBenchmarking.org ms, Fewer Is Better NCNN 20241226 Target: Vulkan GPU - Model: resnet18 NVIDIA 5090 rtx 5090 GeForce RTX 5090 3 6 9 12 15 9.41 10.93 11.01 MIN: 4.47 / MAX: 43.1 MIN: 4.48 / MAX: 44.35 MIN: 4.51 / MAX: 42.89 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: blazeface OpenBenchmarking.org ms, Fewer Is Better NCNN 20241226 Target: Vulkan GPU - Model: blazeface rtx 5090 GeForce RTX 5090 NVIDIA 5090 0.6863 1.3726 2.0589 2.7452 3.4315 2.67 2.67 3.05 MIN: 2.4 / MAX: 41.21 MIN: 2.38 / MAX: 25.86 MIN: 2.38 / MAX: 49.95 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3 OpenBenchmarking.org ms, Fewer Is Better NCNN 20241226 Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3 NVIDIA 5090 GeForce RTX 5090 rtx 5090 1.053 2.106 3.159 4.212 5.265 4.28 4.29 4.68 MIN: 4.05 / MAX: 5.15 MIN: 4.06 / MAX: 5.93 MIN: 4.08 / MAX: 57.94 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPUv2-yolov3v2-yolov3 - Model: mobilenetv2-yolov3 OpenBenchmarking.org ms, Fewer Is Better NCNN 20241226 Target: Vulkan GPUv2-yolov3v2-yolov3 - Model: mobilenetv2-yolov3 GeForce RTX 5090 NVIDIA 5090 rtx 5090 10 20 30 40 50 38.91 39.74 42.44 MIN: 8.9 / MAX: 75.48 MIN: 8.16 / MAX: 76.72 MIN: 8.34 / MAX: 76.17 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: mobilenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20241226 Target: Vulkan GPU - Model: mobilenet GeForce RTX 5090 NVIDIA 5090 rtx 5090 10 20 30 40 50 38.91 39.74 42.44 MIN: 8.9 / MAX: 75.48 MIN: 8.16 / MAX: 76.72 MIN: 8.34 / MAX: 76.17 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: yolov4-tiny OpenBenchmarking.org ms, Fewer Is Better NCNN 20241226 Target: Vulkan GPU - Model: yolov4-tiny GeForce RTX 5090 NVIDIA 5090 rtx 5090 9 18 27 36 45 36.86 38.91 39.34 MIN: 11.12 / MAX: 47.83 MIN: 15.12 / MAX: 48.75 MIN: 15.92 / MAX: 49.04 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: resnet50 OpenBenchmarking.org ms, Fewer Is Better NCNN 20241226 Target: Vulkan GPU - Model: resnet50 rtx 5090 GeForce RTX 5090 NVIDIA 5090 7 14 21 28 35 28.58 29.47 30.10 MIN: 10.02 / MAX: 89.48 MIN: 10.07 / MAX: 90.59 MIN: 10.13 / MAX: 90.77 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
RealSR-NCNN Scale: 4x - TAA: No OpenBenchmarking.org Seconds, Fewer Is Better RealSR-NCNN 20200818 Scale: 4x - TAA: No NVIDIA 5090 GeForce RTX 5090 rtx 5090 1.0418 2.0836 3.1254 4.1672 5.209 4.400 4.446 4.630
NCNN Target: Vulkan GPU - Model: alexnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20241226 Target: Vulkan GPU - Model: alexnet NVIDIA 5090 rtx 5090 GeForce RTX 5090 3 6 9 12 15 8.55 8.74 8.99 MIN: 3.18 / MAX: 21.75 MIN: 3.21 / MAX: 22.23 MIN: 3.19 / MAX: 21.78 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: vgg16 OpenBenchmarking.org ms, Fewer Is Better NCNN 20241226 Target: Vulkan GPU - Model: vgg16 rtx 5090 GeForce RTX 5090 NVIDIA 5090 9 18 27 36 45 37.05 37.49 38.73 MIN: 22.53 / MAX: 46.22 MIN: 23.61 / MAX: 45.51 MIN: 20.99 / MAX: 46.14 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NAMD CUDA ATPase Simulation - 327,506 Atoms OpenBenchmarking.org days/ns, Fewer Is Better NAMD CUDA 2.14 ATPase Simulation - 327,506 Atoms rtx 5090 NVIDIA 5090 GeForce RTX 5090 0.0134 0.0268 0.0402 0.0536 0.067 0.05810 0.05851 0.05943
Blender Blend File: BMW27 - Compute: NVIDIA OptiX OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.3 Blend File: BMW27 - Compute: NVIDIA OptiX rtx 5090 NVIDIA 5090 GeForce RTX 5090 0.6705 1.341 2.0115 2.682 3.3525 2.92 2.97 2.98
SHOC Scalable HeterOgeneous Computing Target: OpenCL - Benchmark: Bus Speed Readback OpenBenchmarking.org GB/s, More Is Better SHOC Scalable HeterOgeneous Computing 2020-04-17 Target: OpenCL - Benchmark: Bus Speed Readback rtx 5090 NVIDIA 5090 GeForce RTX 5090 7 14 21 28 35 28.69 28.59 28.14 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
VkFFT Test: FFT + iFFT C2C 1D batched in half precision OpenBenchmarking.org Benchmark Score, More Is Better VkFFT 1.3.4 Test: FFT + iFFT C2C 1D batched in half precision NVIDIA 5090 rtx 5090 GeForce RTX 5090 70K 140K 210K 280K 350K 305671 302221 300737 1. (CXX) g++ options: -O3
VkFFT Test: FFT + iFFT C2C 1D batched in single precision OpenBenchmarking.org Benchmark Score, More Is Better VkFFT 1.3.4 Test: FFT + iFFT C2C 1D batched in single precision GeForce RTX 5090 rtx 5090 NVIDIA 5090 50K 100K 150K 200K 250K 239575 237717 235884 1. (CXX) g++ options: -O3
VkFFT Test: FFT + iFFT C2C 1D batched in double precision OpenBenchmarking.org Benchmark Score, More Is Better VkFFT 1.3.4 Test: FFT + iFFT C2C 1D batched in double precision rtx 5090 GeForce RTX 5090 NVIDIA 5090 14K 28K 42K 56K 70K 63738 63637 62773 1. (CXX) g++ options: -O3
Blender Blend File: BMW27 - Compute: NVIDIA CUDA OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.3 Blend File: BMW27 - Compute: NVIDIA CUDA GeForce RTX 5090 rtx 5090 NVIDIA 5090 1.062 2.124 3.186 4.248 5.31 4.65 4.72 4.72
Waifu2x-NCNN Vulkan Scale: 2x - Denoise: 3 - TAA: Yes OpenBenchmarking.org Seconds, Fewer Is Better Waifu2x-NCNN Vulkan 20200818 Scale: 2x - Denoise: 3 - TAA: Yes NVIDIA 5090 GeForce RTX 5090 rtx 5090 0.5173 1.0346 1.5519 2.0692 2.5865 2.267 2.278 2.299
Blender Blend File: Barbershop - Compute: NVIDIA OptiX OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.3 Blend File: Barbershop - Compute: NVIDIA OptiX rtx 5090 GeForce RTX 5090 NVIDIA 5090 6 12 18 24 30 24.33 24.50 24.60
Blender Blend File: Pabellon Barcelona - Compute: NVIDIA OptiX OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.3 Blend File: Pabellon Barcelona - Compute: NVIDIA OptiX rtx 5090 GeForce RTX 5090 NVIDIA 5090 2 4 6 8 10 7.00 7.04 7.06
VkFFT Test: FFT + iFFT C2C 1D batched in single precision, no reshuffling OpenBenchmarking.org Benchmark Score, More Is Better VkFFT 1.3.4 Test: FFT + iFFT C2C 1D batched in single precision, no reshuffling rtx 5090 NVIDIA 5090 GeForce RTX 5090 50K 100K 150K 200K 250K 243937 243913 241913 1. (CXX) g++ options: -O3
clpeak OpenCL Test: Transfer Bandwidth enqueueReadBuffer OpenBenchmarking.org GBPS, More Is Better clpeak 1.1.2 OpenCL Test: Transfer Bandwidth enqueueReadBuffer NVIDIA 5090 rtx 5090 GeForce RTX 5090 4 8 12 16 20 13.89 13.83 13.78 1. (CXX) g++ options: -O3
Chaos Group V-RAY Mode: NVIDIA CUDA GPU OpenBenchmarking.org vpaths, More Is Better Chaos Group V-RAY 6.0 Mode: NVIDIA CUDA GPU GeForce RTX 5090 NVIDIA 5090 rtx 5090 1000 2000 3000 4000 5000 4882 4882 4851
Hashcat Benchmark: MD5 OpenBenchmarking.org H/s, More Is Better Hashcat 6.2.4 Benchmark: MD5 rtx 5090 GeForce RTX 5090 NVIDIA 5090 20000M 40000M 60000M 80000M 100000M SE +/- 102551750000.00, N = 2 SE +/- 102256000000.00, N = 2 SE +/- 101883450000.00, N = 2 106848250000 106544000000 106216550000
Blender Blend File: Pabellon Barcelona - Compute: NVIDIA CUDA OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.3 Blend File: Pabellon Barcelona - Compute: NVIDIA CUDA NVIDIA 5090 rtx 5090 GeForce RTX 5090 4 8 12 16 20 17.34 17.35 17.44
SHOC Scalable HeterOgeneous Computing Target: OpenCL - Benchmark: FFT SP OpenBenchmarking.org GFLOPS, More Is Better SHOC Scalable HeterOgeneous Computing 2020-04-17 Target: OpenCL - Benchmark: FFT SP NVIDIA 5090 rtx 5090 GeForce RTX 5090 900 1800 2700 3600 4500 4400.85 4398.39 4375.95 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
Blender Blend File: Barbershop - Compute: NVIDIA CUDA OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.3 Blend File: Barbershop - Compute: NVIDIA CUDA GeForce RTX 5090 rtx 5090 NVIDIA 5090 8 16 24 32 40 35.10 35.14 35.28
clpeak OpenCL Test: Transfer Bandwidth enqueueWriteBuffer OpenBenchmarking.org GBPS, More Is Better clpeak 1.1.2 OpenCL Test: Transfer Bandwidth enqueueWriteBuffer NVIDIA 5090 rtx 5090 GeForce RTX 5090 5 10 15 20 25 18.42 18.41 18.33 1. (CXX) g++ options: -O3
Blender Blend File: Classroom - Compute: NVIDIA OptiX OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.3 Blend File: Classroom - Compute: NVIDIA OptiX NVIDIA 5090 rtx 5090 GeForce RTX 5090 2 4 6 8 10 6.14 6.16 6.17
ProjectPhysX OpenCL-Benchmark Operation: Memory Bandwidth Coalesced Read OpenBenchmarking.org GB/s, More Is Better ProjectPhysX OpenCL-Benchmark 1.6 Operation: Memory Bandwidth Coalesced Read NVIDIA 5090 GeForce RTX 5090 rtx 5090 300 600 900 1200 1500 1603.93 1596.88 1596.24 1. (CXX) g++ options: -std=c++17 -pthread -lOpenCL
ProjectPhysX OpenCL-Benchmark Operation: Memory Bandwidth Coalesced Write OpenBenchmarking.org GB/s, More Is Better ProjectPhysX OpenCL-Benchmark 1.6 Operation: Memory Bandwidth Coalesced Write rtx 5090 GeForce RTX 5090 NVIDIA 5090 400 800 1200 1600 2000 1687.49 1680.23 1679.44 1. (CXX) g++ options: -std=c++17 -pthread -lOpenCL
IndigoBench Acceleration: OpenCL GPU - Scene: Bedroom OpenBenchmarking.org M samples/s, More Is Better IndigoBench 4.4 Acceleration: OpenCL GPU - Scene: Bedroom GeForce RTX 5090 NVIDIA 5090 rtx 5090 10 20 30 40 50 42.92 42.85 42.76
Hashcat Benchmark: 7-Zip OpenBenchmarking.org H/s, More Is Better Hashcat 6.2.4 Benchmark: 7-Zip NVIDIA 5090 rtx 5090 GeForce RTX 5090 700K 1400K 2100K 2800K 3500K 3276600 3272300 3264300
Hashcat Benchmark: SHA1 OpenBenchmarking.org H/s, More Is Better Hashcat 6.2.4 Benchmark: SHA1 GeForce RTX 5090 NVIDIA 5090 rtx 5090 15000M 30000M 45000M 60000M 75000M 69104300000 69072700000 68852500000
SHOC Scalable HeterOgeneous Computing Target: OpenCL - Benchmark: Triad OpenBenchmarking.org GB/s, More Is Better SHOC Scalable HeterOgeneous Computing 2020-04-17 Target: OpenCL - Benchmark: Triad rtx 5090 NVIDIA 5090 GeForce RTX 5090 7 14 21 28 35 27.83 27.81 27.75 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
SHOC Scalable HeterOgeneous Computing Target: OpenCL - Benchmark: S3D OpenBenchmarking.org GFLOPS, More Is Better SHOC Scalable HeterOgeneous Computing 2020-04-17 Target: OpenCL - Benchmark: S3D GeForce RTX 5090 NVIDIA 5090 rtx 5090 200 400 600 800 1000 1120.88 1120.51 1117.54 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
VkFFT Test: FFT + iFFT C2C Bluestein in single precision OpenBenchmarking.org Benchmark Score, More Is Better VkFFT 1.3.4 Test: FFT + iFFT C2C Bluestein in single precision rtx 5090 NVIDIA 5090 GeForce RTX 5090 8K 16K 24K 32K 40K 36054 36026 35954 1. (CXX) g++ options: -O3
Hashcat Benchmark: TrueCrypt RIPEMD160 + XTS OpenBenchmarking.org H/s, More Is Better Hashcat 6.2.4 Benchmark: TrueCrypt RIPEMD160 + XTS GeForce RTX 5090 rtx 5090 NVIDIA 5090 600K 1200K 1800K 2400K 3000K 2777600 2776000 2770700
Blender Blend File: Junkshop - Compute: NVIDIA CUDA OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.3 Blend File: Junkshop - Compute: NVIDIA CUDA NVIDIA 5090 rtx 5090 GeForce RTX 5090 3 6 9 12 15 8.98 8.99 9.00
SHOC Scalable HeterOgeneous Computing Target: OpenCL - Benchmark: GEMM SGEMM_N OpenBenchmarking.org GFLOPS, More Is Better SHOC Scalable HeterOgeneous Computing 2020-04-17 Target: OpenCL - Benchmark: GEMM SGEMM_N NVIDIA 5090 GeForce RTX 5090 rtx 5090 8K 16K 24K 32K 40K 36016.5 35961.3 35937.2 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
Blender Blend File: Fishy Cat - Compute: NVIDIA OptiX OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.3 Blend File: Fishy Cat - Compute: NVIDIA OptiX rtx 5090 GeForce RTX 5090 NVIDIA 5090 1.026 2.052 3.078 4.104 5.13 4.55 4.55 4.56
IndigoBench Acceleration: OpenCL GPU - Scene: Supercar OpenBenchmarking.org M samples/s, More Is Better IndigoBench 4.4 Acceleration: OpenCL GPU - Scene: Supercar GeForce RTX 5090 rtx 5090 NVIDIA 5090 20 40 60 80 100 92.71 92.70 92.52
SHOC Scalable HeterOgeneous Computing Target: OpenCL - Benchmark: MD5 Hash OpenBenchmarking.org GHash/s, More Is Better SHOC Scalable HeterOgeneous Computing 2020-04-17 Target: OpenCL - Benchmark: MD5 Hash GeForce RTX 5090 rtx 5090 NVIDIA 5090 30 60 90 120 150 142.53 142.41 142.24 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
clpeak OpenCL Test: Kernel Latency OpenBenchmarking.org us, Fewer Is Better clpeak 1.1.2 OpenCL Test: Kernel Latency rtx 5090 GeForce RTX 5090 NVIDIA 5090 1.161 2.322 3.483 4.644 5.805 5.15 5.15 5.16 1. (CXX) g++ options: -O3
ProjectPhysX OpenCL-Benchmark Operation: INT64 Compute OpenBenchmarking.org TIOPs/s, More Is Better ProjectPhysX OpenCL-Benchmark 1.6 Operation: INT64 Compute NVIDIA 5090 rtx 5090 GeForce RTX 5090 0.99 1.98 2.97 3.96 4.95 4.400 4.396 4.392 1. (CXX) g++ options: -std=c++17 -pthread -lOpenCL
Blender Blend File: Junkshop - Compute: NVIDIA OptiX OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.3 Blend File: Junkshop - Compute: NVIDIA OptiX rtx 5090 GeForce RTX 5090 NVIDIA 5090 1.2758 2.5516 3.8274 5.1032 6.379 5.66 5.66 5.67
ProjectPhysX OpenCL-Benchmark Operation: INT8 Compute OpenBenchmarking.org TIOPs/s, More Is Better ProjectPhysX OpenCL-Benchmark 1.6 Operation: INT8 Compute rtx 5090 GeForce RTX 5090 NVIDIA 5090 10 20 30 40 50 41.80 41.76 41.72 1. (CXX) g++ options: -std=c++17 -pthread -lOpenCL
SHOC Scalable HeterOgeneous Computing Target: OpenCL - Benchmark: Texture Read Bandwidth OpenBenchmarking.org GB/s, More Is Better SHOC Scalable HeterOgeneous Computing 2020-04-17 Target: OpenCL - Benchmark: Texture Read Bandwidth GeForce RTX 5090 NVIDIA 5090 rtx 5090 600 1200 1800 2400 3000 2875.50 2872.53 2870.68 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
NCNN Target: Vulkan GPU - Model: vision_transformer OpenBenchmarking.org ms, Fewer Is Better NCNN 20241226 Target: Vulkan GPU - Model: vision_transformer rtx 5090 GeForce RTX 5090 NVIDIA 5090 14 28 42 56 70 62.76 62.83 62.86 MIN: 40.3 / MAX: 105.61 MIN: 41.21 / MAX: 109.15 MIN: 42.12 / MAX: 106.46 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
ProjectPhysX OpenCL-Benchmark Operation: INT16 Compute OpenBenchmarking.org TIOPs/s, More Is Better ProjectPhysX OpenCL-Benchmark 1.6 Operation: INT16 Compute NVIDIA 5090 rtx 5090 GeForce RTX 5090 12 24 36 48 60 54.04 54.02 53.95 1. (CXX) g++ options: -std=c++17 -pthread -lOpenCL
RealSR-NCNN Scale: 4x - TAA: Yes OpenBenchmarking.org Seconds, Fewer Is Better RealSR-NCNN 20200818 Scale: 4x - TAA: Yes rtx 5090 GeForce RTX 5090 NVIDIA 5090 3 6 9 12 15 13.48 13.49 13.50
clpeak OpenCL Test: Global Memory Bandwidth OpenBenchmarking.org GBPS, More Is Better clpeak 1.1.2 OpenCL Test: Global Memory Bandwidth GeForce RTX 5090 NVIDIA 5090 rtx 5090 300 600 900 1200 1500 1564.61 1564.49 1562.97 1. (CXX) g++ options: -O3
FluidX3D Test: FP32-FP16C OpenBenchmarking.org MLUPs/s, More Is Better FluidX3D 3.0 Test: FP32-FP16C rtx 5090 GeForce RTX 5090 NVIDIA 5090 4K 8K 12K 16K 20K 19140 19135 19121
clpeak OpenCL Test: Integer 24-bit Compute OpenBenchmarking.org GIOPS, More Is Better clpeak 1.1.2 OpenCL Test: Integer 24-bit Compute GeForce RTX 5090 NVIDIA 5090 rtx 5090 13K 26K 39K 52K 65K 61903.93 61866.86 61843.11 1. (CXX) g++ options: -O3
clpeak OpenCL Test: Integer Compute OpenBenchmarking.org GIOPS, More Is Better clpeak 1.1.2 OpenCL Test: Integer Compute GeForce RTX 5090 rtx 5090 NVIDIA 5090 13K 26K 39K 52K 65K 62178.05 62151.94 62119.51 1. (CXX) g++ options: -O3
VkFFT Test: FFT + iFFT C2C Bluestein benchmark in double precision OpenBenchmarking.org Benchmark Score, More Is Better VkFFT 1.3.4 Test: FFT + iFFT C2C Bluestein benchmark in double precision NVIDIA 5090 GeForce RTX 5090 rtx 5090 2K 4K 6K 8K 10K 9940 9934 9931 1. (CXX) g++ options: -O3
VkFFT Test: FFT + iFFT R2C / C2R OpenBenchmarking.org Benchmark Score, More Is Better VkFFT 1.3.4 Test: FFT + iFFT R2C / C2R rtx 5090 GeForce RTX 5090 NVIDIA 5090 40K 80K 120K 160K 200K 164933 164873 164801 1. (CXX) g++ options: -O3
SHOC Scalable HeterOgeneous Computing Target: OpenCL - Benchmark: Max SP Flops OpenBenchmarking.org GFLOPS, More Is Better SHOC Scalable HeterOgeneous Computing 2020-04-17 Target: OpenCL - Benchmark: Max SP Flops GeForce RTX 5090 rtx 5090 NVIDIA 5090 30K 60K 90K 120K 150K 124646 124615 124556 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
VkFFT Test: FFT + iFFT C2C multidimensional in single precision OpenBenchmarking.org Benchmark Score, More Is Better VkFFT 1.3.4 Test: FFT + iFFT C2C multidimensional in single precision GeForce RTX 5090 rtx 5090 NVIDIA 5090 30K 60K 90K 120K 150K 144695 144624 144600 1. (CXX) g++ options: -O3
Hashcat Benchmark: SHA-512 OpenBenchmarking.org H/s, More Is Better Hashcat 6.2.4 Benchmark: SHA-512 NVIDIA 5090 rtx 5090 GeForce RTX 5090 2000M 4000M 6000M 8000M 10000M 8901200000 8900400000 8895900000
ProjectPhysX OpenCL-Benchmark Operation: FP64 Compute OpenBenchmarking.org TFLOPs/s, More Is Better ProjectPhysX OpenCL-Benchmark 1.6 Operation: FP64 Compute GeForce RTX 5090 NVIDIA 5090 rtx 5090 0.439 0.878 1.317 1.756 2.195 1.951 1.950 1.950 1. (CXX) g++ options: -std=c++17 -pthread -lOpenCL
VkResample Upscale: 2x - Precision: Double OpenBenchmarking.org ms, Fewer Is Better VkResample 1.0 Upscale: 2x - Precision: Double GeForce RTX 5090 NVIDIA 5090 rtx 5090 20 40 60 80 100 103.45 103.48 103.51 1. (CXX) g++ options: -O3
vkpeak fp32-vec4 OpenBenchmarking.org GFLOPS, More Is Better vkpeak 20240505 fp32-vec4 rtx 5090 NVIDIA 5090 GeForce RTX 5090 20K 40K 60K 80K 100K 83296.84 83290.22 83257.59
vkpeak int32-vec4 OpenBenchmarking.org GIOPS, More Is Better vkpeak 20240505 int32-vec4 GeForce RTX 5090 NVIDIA 5090 rtx 5090 13K 26K 39K 52K 65K 61914.51 61894.95 61885.37
vkpeak int16-scalar OpenBenchmarking.org GIOPS, More Is Better vkpeak 20240505 int16-scalar rtx 5090 GeForce RTX 5090 NVIDIA 5090 9K 18K 27K 36K 45K 40006.60 39998.42 39989.70
vkpeak fp32-scalar OpenBenchmarking.org GFLOPS, More Is Better vkpeak 20240505 fp32-scalar GeForce RTX 5090 NVIDIA 5090 rtx 5090 14K 28K 42K 56K 70K 63035.62 63035.62 63013.32
FluidX3D Test: FP32-FP32 OpenBenchmarking.org MLUPs/s, More Is Better FluidX3D 3.0 Test: FP32-FP32 NVIDIA 5090 GeForce RTX 5090 rtx 5090 2K 4K 6K 8K 10K 9527 9525 9524
SHOC Scalable HeterOgeneous Computing Target: OpenCL - Benchmark: Reduction OpenBenchmarking.org GB/s, More Is Better SHOC Scalable HeterOgeneous Computing 2020-04-17 Target: OpenCL - Benchmark: Reduction GeForce RTX 5090 rtx 5090 NVIDIA 5090 200 400 600 800 1000 837.24 837.21 836.98 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
ProjectPhysX OpenCL-Benchmark Operation: FP32 Compute OpenBenchmarking.org TFLOPs/s, More Is Better ProjectPhysX OpenCL-Benchmark 1.6 Operation: FP32 Compute GeForce RTX 5090 NVIDIA 5090 rtx 5090 30 60 90 120 150 117.88 117.86 117.85 1. (CXX) g++ options: -std=c++17 -pthread -lOpenCL
ProjectPhysX OpenCL-Benchmark Operation: FP16 Compute OpenBenchmarking.org TFLOPs/s, More Is Better ProjectPhysX OpenCL-Benchmark 1.6 Operation: FP16 Compute NVIDIA 5090 GeForce RTX 5090 rtx 5090 30 60 90 120 150 122.94 122.94 122.91 1. (CXX) g++ options: -std=c++17 -pthread -lOpenCL
clpeak OpenCL Test: Double-Precision Compute OpenBenchmarking.org GFLOPS, More Is Better clpeak 1.1.2 OpenCL Test: Double-Precision Compute NVIDIA 5090 rtx 5090 GeForce RTX 5090 400 800 1200 1600 2000 1977.26 1976.90 1976.78 1. (CXX) g++ options: -O3
vkpeak fp16-vec4 OpenBenchmarking.org GFLOPS, More Is Better vkpeak 20240505 fp16-vec4 rtx 5090 NVIDIA 5090 GeForce RTX 5090 16K 32K 48K 64K 80K 72592.93 72578.62 72575.33
vkpeak fp16-scalar OpenBenchmarking.org GFLOPS, More Is Better vkpeak 20240505 fp16-scalar rtx 5090 NVIDIA 5090 GeForce RTX 5090 13K 26K 39K 52K 65K 62611.78 62611.17 62597.51
ProjectPhysX OpenCL-Benchmark Operation: INT32 Compute OpenBenchmarking.org TIOPs/s, More Is Better ProjectPhysX OpenCL-Benchmark 1.6 Operation: INT32 Compute GeForce RTX 5090 NVIDIA 5090 rtx 5090 14 28 42 56 70 61.77 61.76 61.76 1. (CXX) g++ options: -std=c++17 -pthread -lOpenCL
FluidX3D Test: FP32-FP16S OpenBenchmarking.org MLUPs/s, More Is Better FluidX3D 3.0 Test: FP32-FP16S GeForce RTX 5090 rtx 5090 NVIDIA 5090 4K 8K 12K 16K 20K 18500 18499 18496
vkpeak fp64-vec4 OpenBenchmarking.org GFLOPS, More Is Better vkpeak 20240505 fp64-vec4 rtx 5090 NVIDIA 5090 GeForce RTX 5090 400 800 1200 1600 2000 1965.70 1965.39 1965.32
clpeak OpenCL Test: Single-Precision Compute OpenBenchmarking.org GFLOPS, More Is Better clpeak 1.1.2 OpenCL Test: Single-Precision Compute NVIDIA 5090 GeForce RTX 5090 rtx 5090 30K 60K 90K 120K 150K 121438.41 121419.57 121415.53 1. (CXX) g++ options: -O3
VkResample Upscale: 2x - Precision: Single OpenBenchmarking.org ms, Fewer Is Better VkResample 1.0 Upscale: 2x - Precision: Single rtx 5090 GeForce RTX 5090 NVIDIA 5090 1.271 2.542 3.813 5.084 6.355 5.648 5.648 5.649 1. (CXX) g++ options: -O3
vkpeak int16-vec4 OpenBenchmarking.org GIOPS, More Is Better vkpeak 20240505 int16-vec4 rtx 5090 GeForce RTX 5090 NVIDIA 5090 9K 18K 27K 36K 45K 43806.13 43803.60 43799.96
vkpeak fp64-scalar OpenBenchmarking.org GFLOPS, More Is Better vkpeak 20240505 fp64-scalar NVIDIA 5090 GeForce RTX 5090 rtx 5090 400 800 1200 1600 2000 1967.49 1967.43 1967.37
SHOC Scalable HeterOgeneous Computing Target: OpenCL - Benchmark: Bus Speed Download OpenBenchmarking.org GB/s, More Is Better SHOC Scalable HeterOgeneous Computing 2020-04-17 Target: OpenCL - Benchmark: Bus Speed Download GeForce RTX 5090 NVIDIA 5090 rtx 5090 7 14 21 28 35 28.79 28.79 28.79 1. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
vkpeak int32-scalar OpenBenchmarking.org GIOPS, More Is Better vkpeak 20240505 int32-scalar rtx 5090 NVIDIA 5090 GeForce RTX 5090 13K 26K 39K 52K 65K 62142.56 62142.07 62141.02
Chaos Group V-RAY Mode: NVIDIA RTX GPU OpenBenchmarking.org vpaths, More Is Better Chaos Group V-RAY 6.0 Mode: NVIDIA RTX GPU GeForce RTX 5090 NVIDIA 5090 rtx 5090 3K 6K 9K 12K 15K 11923 11923 11923
Blender Blend File: Fishy Cat - Compute: NVIDIA CUDA OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.3 Blend File: Fishy Cat - Compute: NVIDIA CUDA rtx 5090 NVIDIA 5090 GeForce RTX 5090 2 4 6 8 10 8.92 8.92 8.92
Blender Blend File: Classroom - Compute: NVIDIA CUDA OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.3 Blend File: Classroom - Compute: NVIDIA CUDA rtx 5090 NVIDIA 5090 GeForce RTX 5090 2 4 6 8 10 8.38 8.38 8.38
Phoronix Test Suite v10.8.5