ncnn llama

Intel Core Ultra 7 256V testing with a ASUS Zenbook S 14 UX5406SA_UX5406SA UX5406SA v1.0 (UX5406SA.300 BIOS) and ASUS Intel LNL 7GB on Ubuntu 24.10 via the Phoronix Test Suite.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2412296-NE-NCNNLLAMA15
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results
Show Result Confidence Charts
Allow Limiting Results To Certain Suite(s)

Statistics

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

Graph Settings

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

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Toggle/Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
a
December 29 2024
  1 Hour, 53 Minutes
b
December 29 2024
  1 Hour, 48 Minutes
c
December 29 2024
  1 Hour, 49 Minutes
d
December 29 2024
  1 Hour, 49 Minutes
Invert Behavior (Only Show Selected Data)
  1 Hour, 50 Minutes

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


ncnn llamaOpenBenchmarking.orgPhoronix Test SuiteIntel Core Ultra 7 256V @ 4.70GHz (8 Cores)ASUS Zenbook S 14 UX5406SA_UX5406SA UX5406SA v1.0 (UX5406SA.300 BIOS)Intel Device a87f8 x 2GB LPDDR5-8533MT/s Samsung1024GB Western Digital WD PC SN560 SDDPNQE-1T00-1102ASUS Intel LNL 7GBIntel Lunar Lake-M HD AudioIntel Device a840Ubuntu 24.106.12.0-rc6-phx-drm-next (x86_64)GNOME Shell 47.0X Server + Wayland4.6 Mesa 25.0~git2411250600.45c523~oibaf~o (git-45c5231 2024-11-25 oracular-oibaf-ppOpenCL 3.0GCC 14.2.0ext42880x1800ProcessorMotherboardChipsetMemoryDiskGraphicsAudioNetworkOSKernelDesktopDisplay ServerOpenGLOpenCLCompilerFile-SystemScreen ResolutionNcnn Llama BenchmarksSystem Logs- Transparent Huge Pages: madvise- --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 - Scaling Governor: intel_pstate powersave (EPP: performance) - Platform Profile: performance - CPU Microcode: 0x114 - Thermald 2.5.8 - ACPI Profile: performance - 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

abcdResult OverviewPhoronix Test Suite100%115%130%146%NCNNNCNNNCNNNCNNNCNNNCNNNCNNNCNNNCNNNCNNNCNNNCNNLlama.cppNCNNLlama.cppNCNNLlama.cppNCNNNCNNNCNNNCNNNCNNNCNNLlama.cppNCNNNCNNNCNNNCNNLlama.cppNCNNNCNNLlama.cppNCNNNCNNNCNNLlama.cppNCNNLlama.cppNCNNNCNNNCNNLlama.cppLlama.cppNCNNLlama.cppLlama.cppNCNNNCNNCPU - resnet50CPU - squeezenet_ssdCPU - resnet18CPU - googlenetCPU - alexnetCPU-v2-v2 - mobilenet-v2CPU - efficientnet-b0CPU - mnasnetCPU-v3-v3 - mobilenet-v3CPU - shufflenet-v2CPU - regnety_400mCPU - blazefaceCPU BLAS - granite-3.0-3b-a800m-instruct-Q8_0 - P.P.1CPU - vgg16CPU BLAS - granite-3.0-3b-a800m-instruct-Q8_0 - P.P.2Vulkan GPU - FastestDetCPU BLAS - Mistral-7B-Instruct-v0.3-Q8_0 - P.P.1CPU - mobilenetCPUv2-yolov3v2-yolov3 - mobilenetv2-yolov3CPU - FastestDetVulkan GPU - mobilenetV.G.y.y - mobilenetv2-yolov3Vulkan GPU-v2-v2 - mobilenet-v2CPU BLAS - Llama-3.1-Tulu-3-8B-Q8_0 - P.P.1Vulkan GPU - alexnetVulkan GPU-v3-v3 - mobilenet-v3Vulkan GPU - regnety_400mVulkan GPU - vision_transformerCPU BLAS - granite-3.0-3b-a800m-instruct-Q8_0 - P.P.5Vulkan GPU - mnasnetCPU - yolov4-tinyCPU BLAS - granite-3.0-3b-a800m-instruct-Q8_0 - T.G.1Vulkan GPU - squeezenet_ssdVulkan GPU - yolov4-tinyVulkan GPU - efficientnet-b0CPU BLAS - Mistral-7B-Instruct-v0.3-Q8_0 - P.P.5Vulkan GPU - resnet18CPU BLAS - Llama-3.1-Tulu-3-8B-Q8_0 - P.P.5CPU - vision_transformerVulkan GPU - resnet50Vulkan GPU - vgg16CPU BLAS - Llama-3.1-Tulu-3-8B-Q8_0 - P.P.2CPU BLAS - Mistral-7B-Instruct-v0.3-Q8_0 - P.P.2Vulkan GPU - blazefaceCPU BLAS - Llama-3.1-Tulu-3-8B-Q8_0 - T.G.1CPU BLAS - Mistral-7B-Instruct-v0.3-Q8_0 - T.G.1Vulkan GPU - shufflenet-v2Vulkan GPU - googlenet

ncnn llamancnn: CPU - mobilenetncnn: CPU-v2-v2 - mobilenet-v2ncnn: CPU-v3-v3 - mobilenet-v3ncnn: CPU - shufflenet-v2ncnn: CPU - mnasnetncnn: CPU - efficientnet-b0ncnn: CPU - blazefacencnn: CPU - googlenetncnn: CPU - vgg16ncnn: CPU - resnet18ncnn: CPU - alexnetncnn: CPU - resnet50ncnn: CPUv2-yolov3v2-yolov3 - mobilenetv2-yolov3ncnn: CPU - yolov4-tinyncnn: CPU - squeezenet_ssdncnn: CPU - regnety_400mncnn: CPU - vision_transformerncnn: CPU - FastestDetncnn: Vulkan GPU - mobilenetncnn: Vulkan GPU-v2-v2 - mobilenet-v2ncnn: Vulkan GPU-v3-v3 - mobilenet-v3ncnn: Vulkan GPU - shufflenet-v2ncnn: Vulkan GPU - mnasnetncnn: Vulkan GPU - efficientnet-b0ncnn: Vulkan GPU - blazefacencnn: Vulkan GPU - googlenetncnn: Vulkan GPU - vgg16ncnn: Vulkan GPU - resnet18ncnn: Vulkan GPU - alexnetncnn: Vulkan GPU - resnet50ncnn: Vulkan GPUv2-yolov3v2-yolov3 - mobilenetv2-yolov3ncnn: Vulkan GPU - yolov4-tinyncnn: Vulkan GPU - squeezenet_ssdncnn: Vulkan GPU - regnety_400mncnn: Vulkan GPU - vision_transformerncnn: Vulkan GPU - FastestDetllama-cpp: CPU BLAS - Llama-3.1-Tulu-3-8B-Q8_0 - Text Generation 128llama-cpp: CPU BLAS - Llama-3.1-Tulu-3-8B-Q8_0 - Prompt Processing 512llama-cpp: CPU BLAS - Llama-3.1-Tulu-3-8B-Q8_0 - Prompt Processing 1024llama-cpp: CPU BLAS - Llama-3.1-Tulu-3-8B-Q8_0 - Prompt Processing 2048llama-cpp: CPU BLAS - Mistral-7B-Instruct-v0.3-Q8_0 - Text Generation 128llama-cpp: CPU BLAS - Mistral-7B-Instruct-v0.3-Q8_0 - Prompt Processing 512llama-cpp: CPU BLAS - Mistral-7B-Instruct-v0.3-Q8_0 - Prompt Processing 1024llama-cpp: CPU BLAS - Mistral-7B-Instruct-v0.3-Q8_0 - Prompt Processing 2048llama-cpp: CPU BLAS - granite-3.0-3b-a800m-instruct-Q8_0 - Text Generation 128llama-cpp: CPU BLAS - granite-3.0-3b-a800m-instruct-Q8_0 - Prompt Processing 512llama-cpp: CPU BLAS - granite-3.0-3b-a800m-instruct-Q8_0 - Prompt Processing 1024llama-cpp: CPU BLAS - granite-3.0-3b-a800m-instruct-Q8_0 - Prompt Processing 2048abcd14.147.016.134.826.3311.162.3716.6248.7211.368.3930.4914.1417.715.1721.56199.024.9313.994.754.513.654.477.921.9211.2544.957.565.9119.113.9917.649.5716.59196.624.828.8728.7728.0927.169.2428.72827.1138.3561.7360.8358.7313.5954.53.654.487.991.8411.2144.47.515.8618.9613.5917.629.5516.51199.844.6914.354.874.393.664.377.991.9211.2844.617.656.1719.2614.3517.749.7716.73200.124.958.8528.6127.9327.099.2228.327.927.2238.3561.3461.4961.414.44.94.433.644.497.841.8811.2644.857.596.1219.2514.417.549.6616.9197.494.6813.8854.533.654.47.871.9211.2744.657.555.9119.0613.8817.519.6616.39202.374.818.8528.4327.7827.159.228.5426.2527.1938.4360.8554.0856.7914.294.984.533.654.527.951.9411.3644.327.626.0619.214.2917.939.6316.34199.34.9413.634.964.493.664.427.981.9111.2545.047.565.9419.0613.6317.849.716.25198.734.638.8328.5126.8727.019.2128.527.8127.2637.6462.3154.3957.28OpenBenchmarking.org

NCNN

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: CPU - Model: mobilenetabcd4812162014.1413.5914.4014.29MIN: 12.81 / MAX: 19.39MIN: 12.35 / MAX: 16.17MIN: 13.42 / MAX: 16.32MIN: 13.43 / MAX: 15.381. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: CPU-v2-v2 - Model: mobilenet-v2abcd2468107.015.004.904.98MIN: 6.26 / MAX: 8.27MIN: 4.79 / MAX: 6.66MIN: 4.22 / MAX: 7.14MIN: 4.61 / MAX: 7.091. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: CPU-v3-v3 - Model: mobilenet-v3abcd2468106.134.504.434.53MIN: 5.69 / MAX: 8.86MIN: 4.3 / MAX: 7.07MIN: 4.15 / MAX: 6.44MIN: 4.32 / MAX: 6.541. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: CPU - Model: shufflenet-v2abcd1.08452.1693.25354.3385.42254.823.653.643.65MIN: 4.59 / MAX: 5.02MIN: 3.46 / MAX: 5.16MIN: 3.49 / MAX: 4.97MIN: 3.6 / MAX: 5.031. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: CPU - Model: mnasnetabcd2468106.334.484.494.52MIN: 5.92 / MAX: 7.4MIN: 4.18 / MAX: 4.99MIN: 4.38 / MAX: 5.08MIN: 4.29 / MAX: 7.541. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: CPU - Model: efficientnet-b0abcd369121511.167.997.847.95MIN: 10.59 / MAX: 11.76MIN: 7.54 / MAX: 9.91MIN: 7.34 / MAX: 9.39MIN: 7.4 / MAX: 8.361. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: CPU - Model: blazefaceabcd0.53331.06661.59992.13322.66652.371.841.881.94MIN: 2.23 / MAX: 2.49MIN: 1.74 / MAX: 1.91MIN: 1.82 / MAX: 1.99MIN: 1.9 / MAX: 1.991. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: CPU - Model: googlenetabcd4812162016.6211.2111.2611.36MIN: 15.75 / MAX: 17.72MIN: 10.76 / MAX: 11.55MIN: 10.66 / MAX: 13.28MIN: 11.11 / MAX: 11.761. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: CPU - Model: vgg16abcd112233445548.7244.4044.8544.32MIN: 47.5 / MAX: 50.51MIN: 41.87 / MAX: 45.98MIN: 42.51 / MAX: 46.34MIN: 41.83 / MAX: 45.851. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: CPU - Model: resnet18abcd369121511.367.517.597.62MIN: 10.53 / MAX: 12.29MIN: 7.26 / MAX: 7.95MIN: 7.4 / MAX: 7.96MIN: 7.39 / MAX: 8.421. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: CPU - Model: alexnetabcd2468108.395.866.126.06MIN: 7.52 / MAX: 9.64MIN: 5.71 / MAX: 6.23MIN: 5.95 / MAX: 6.36MIN: 5.91 / MAX: 7.61. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: CPU - Model: resnet50abcd71421283530.4918.9619.2519.20MIN: 28.71 / MAX: 34.17MIN: 18.53 / MAX: 19.8MIN: 18.8 / MAX: 21.77MIN: 18.8 / MAX: 19.921. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: CPUv2-yolov3v2-yolov3 - Model: mobilenetv2-yolov3abcd4812162014.1413.5914.4014.29MIN: 12.81 / MAX: 19.39MIN: 12.35 / MAX: 16.17MIN: 13.42 / MAX: 16.32MIN: 13.43 / MAX: 15.381. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: CPU - Model: yolov4-tinyabcd4812162017.7017.6217.5417.93MIN: 16.34 / MAX: 19.92MIN: 16.86 / MAX: 18.72MIN: 16.19 / MAX: 18.43MIN: 16.96 / MAX: 19.341. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: CPU - Model: squeezenet_ssdabcd4812162015.179.559.669.63MIN: 13.87 / MAX: 20.8MIN: 9.04 / MAX: 10.3MIN: 9.09 / MAX: 10.11MIN: 9.13 / MAX: 10.021. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: CPU - Model: regnety_400mabcd51015202521.5616.5116.9016.34MIN: 20.87 / MAX: 28.32MIN: 15.62 / MAX: 18.22MIN: 15.72 / MAX: 18MIN: 15.56 / MAX: 21.351. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: CPU - Model: vision_transformerabcd4080120160200199.02199.84197.49199.30MIN: 194.44 / MAX: 203.76MIN: 194.4 / MAX: 204.25MIN: 193.52 / MAX: 203.23MIN: 195.1 / MAX: 203.241. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: CPU - Model: FastestDetabcd1.11152.2233.33454.4465.55754.934.694.684.94MIN: 4.85 / MAX: 5.13MIN: 4.36 / MAX: 4.91MIN: 4.45 / MAX: 5.02MIN: 4.53 / MAX: 5.071. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: Vulkan GPU - Model: mobilenetabcd4812162013.9914.3513.8813.63MIN: 12.4 / MAX: 16.02MIN: 13.46 / MAX: 16.3MIN: 12.41 / MAX: 15.7MIN: 12.41 / MAX: 15.261. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2abcd1.1252.253.3754.55.6254.754.875.004.96MIN: 4.23 / MAX: 5.52MIN: 4.22 / MAX: 7.49MIN: 4.26 / MAX: 7.25MIN: 4.56 / MAX: 7.551. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3abcd1.01932.03863.05794.07725.09654.514.394.534.49MIN: 4.25 / MAX: 4.9MIN: 4.16 / MAX: 6.09MIN: 4.3 / MAX: 7.77MIN: 4.21 / MAX: 5.781. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: Vulkan GPU - Model: shufflenet-v2abcd0.82351.6472.47053.2944.11753.653.663.653.66MIN: 3.6 / MAX: 3.72MIN: 3.58 / MAX: 6.33MIN: 3.48 / MAX: 5.38MIN: 3.59 / MAX: 5.531. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: Vulkan GPU - Model: mnasnetabcd1.00582.01163.01744.02325.0294.474.374.404.42MIN: 4.22 / MAX: 5.04MIN: 4.19 / MAX: 4.99MIN: 4.22 / MAX: 4.98MIN: 4.23 / MAX: 4.941. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: Vulkan GPU - Model: efficientnet-b0abcd2468107.927.997.877.98MIN: 7.56 / MAX: 8.16MIN: 7.88 / MAX: 8.15MIN: 7.36 / MAX: 9.34MIN: 7.68 / MAX: 8.291. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: Vulkan GPU - Model: blazefaceabcd0.4320.8641.2961.7282.161.921.921.921.91MIN: 1.87 / MAX: 1.98MIN: 1.83 / MAX: 2.03MIN: 1.84 / MAX: 1.98MIN: 1.84 / MAX: 1.971. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: Vulkan GPU - Model: googlenetabcd369121511.2511.2811.2711.25MIN: 10.92 / MAX: 12.23MIN: 10.66 / MAX: 11.92MIN: 10.72 / MAX: 14.11MIN: 10.97 / MAX: 12.351. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: Vulkan GPU - Model: vgg16abcd102030405044.9544.6144.6545.04MIN: 42.32 / MAX: 46.93MIN: 42.3 / MAX: 46.22MIN: 41.13 / MAX: 46.45MIN: 43.03 / MAX: 46.431. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: Vulkan GPU - Model: resnet18abcd2468107.567.657.557.56MIN: 7.29 / MAX: 7.92MIN: 7.42 / MAX: 7.91MIN: 7.3 / MAX: 9.6MIN: 7.28 / MAX: 7.941. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: Vulkan GPU - Model: alexnetabcd2468105.916.175.915.94MIN: 5.73 / MAX: 6.22MIN: 5.98 / MAX: 6.59MIN: 5.74 / MAX: 6.27MIN: 5.75 / MAX: 6.231. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: Vulkan GPU - Model: resnet50abcd51015202519.1019.2619.0619.06MIN: 18.54 / MAX: 21.48MIN: 18.87 / MAX: 20.2MIN: 18.63 / MAX: 19.69MIN: 18.66 / MAX: 19.881. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: Vulkan GPUv2-yolov3v2-yolov3 - Model: mobilenetv2-yolov3abcd4812162013.9914.3513.8813.63MIN: 12.4 / MAX: 16.02MIN: 13.46 / MAX: 16.3MIN: 12.41 / MAX: 15.7MIN: 12.41 / MAX: 15.261. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: Vulkan GPU - Model: yolov4-tinyabcd4812162017.6417.7417.5117.84MIN: 16.96 / MAX: 18.52MIN: 16.9 / MAX: 19.03MIN: 16.35 / MAX: 19.44MIN: 16.99 / MAX: 19.521. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: Vulkan GPU - Model: squeezenet_ssdabcd36912159.579.779.669.70MIN: 9.13 / MAX: 11.39MIN: 9.06 / MAX: 10.17MIN: 8.92 / MAX: 10.07MIN: 9.2 / MAX: 10.591. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: Vulkan GPU - Model: regnety_400mabcd4812162016.5916.7316.3916.25MIN: 15.66 / MAX: 19.68MIN: 15.69 / MAX: 21MIN: 15.67 / MAX: 19.65MIN: 15.65 / MAX: 17.821. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: Vulkan GPU - Model: vision_transformerabcd4080120160200196.62200.12202.37198.73MIN: 192.65 / MAX: 200.94MIN: 194.52 / MAX: 204.7MIN: 195.84 / MAX: 206.71MIN: 194.73 / MAX: 203.131. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20241226Target: Vulkan GPU - Model: FastestDetabcd1.11382.22763.34144.45525.5694.824.954.814.63MIN: 4.71 / MAX: 5.03MIN: 4.86 / MAX: 5.31MIN: 4.69 / MAX: 5.02MIN: 4.33 / MAX: 4.891. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Llama.cpp

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b4397Backend: CPU BLAS - Model: Llama-3.1-Tulu-3-8B-Q8_0 - Test: Text Generation 128abcd2468108.878.858.858.831. (CXX) g++ options: -O3

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b4397Backend: CPU BLAS - Model: Llama-3.1-Tulu-3-8B-Q8_0 - Test: Prompt Processing 512abcd71421283528.7728.6128.4328.511. (CXX) g++ options: -O3

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b4397Backend: CPU BLAS - Model: Llama-3.1-Tulu-3-8B-Q8_0 - Test: Prompt Processing 1024abcd71421283528.0927.9327.7826.871. (CXX) g++ options: -O3

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b4397Backend: CPU BLAS - Model: Llama-3.1-Tulu-3-8B-Q8_0 - Test: Prompt Processing 2048abcd61218243027.1627.0927.1527.011. (CXX) g++ options: -O3

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b4397Backend: CPU BLAS - Model: Mistral-7B-Instruct-v0.3-Q8_0 - Test: Text Generation 128abcd36912159.249.229.209.211. (CXX) g++ options: -O3

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b4397Backend: CPU BLAS - Model: Mistral-7B-Instruct-v0.3-Q8_0 - Test: Prompt Processing 512abcd71421283528.7028.3028.5428.501. (CXX) g++ options: -O3

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b4397Backend: CPU BLAS - Model: Mistral-7B-Instruct-v0.3-Q8_0 - Test: Prompt Processing 1024abcd71421283528.0027.9026.2527.811. (CXX) g++ options: -O3

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b4397Backend: CPU BLAS - Model: Mistral-7B-Instruct-v0.3-Q8_0 - Test: Prompt Processing 2048abcd61218243027.1127.2227.1927.261. (CXX) g++ options: -O3

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b4397Backend: CPU BLAS - Model: granite-3.0-3b-a800m-instruct-Q8_0 - Test: Text Generation 128abcd91827364538.3538.3538.4337.641. (CXX) g++ options: -O3

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b4397Backend: CPU BLAS - Model: granite-3.0-3b-a800m-instruct-Q8_0 - Test: Prompt Processing 512abcd142842567061.7361.3460.8562.311. (CXX) g++ options: -O3

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b4397Backend: CPU BLAS - Model: granite-3.0-3b-a800m-instruct-Q8_0 - Test: Prompt Processing 1024abcd142842567060.8361.4954.0854.391. (CXX) g++ options: -O3

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b4397Backend: CPU BLAS - Model: granite-3.0-3b-a800m-instruct-Q8_0 - Test: Prompt Processing 2048abcd142842567058.7361.4056.7957.281. (CXX) g++ options: -O3

48 Results Shown

NCNN:
  CPU - mobilenet
  CPU-v2-v2 - mobilenet-v2
  CPU-v3-v3 - mobilenet-v3
  CPU - shufflenet-v2
  CPU - mnasnet
  CPU - efficientnet-b0
  CPU - blazeface
  CPU - googlenet
  CPU - vgg16
  CPU - resnet18
  CPU - alexnet
  CPU - resnet50
  CPUv2-yolov3v2-yolov3 - mobilenetv2-yolov3
  CPU - yolov4-tiny
  CPU - squeezenet_ssd
  CPU - regnety_400m
  CPU - vision_transformer
  CPU - FastestDet
  Vulkan GPU - mobilenet
  Vulkan GPU-v2-v2 - mobilenet-v2
  Vulkan GPU-v3-v3 - mobilenet-v3
  Vulkan GPU - shufflenet-v2
  Vulkan GPU - mnasnet
  Vulkan GPU - efficientnet-b0
  Vulkan GPU - blazeface
  Vulkan GPU - googlenet
  Vulkan GPU - vgg16
  Vulkan GPU - resnet18
  Vulkan GPU - alexnet
  Vulkan GPU - resnet50
  Vulkan GPUv2-yolov3v2-yolov3 - mobilenetv2-yolov3
  Vulkan GPU - yolov4-tiny
  Vulkan GPU - squeezenet_ssd
  Vulkan GPU - regnety_400m
  Vulkan GPU - vision_transformer
  Vulkan GPU - FastestDet
Llama.cpp:
  CPU BLAS - Llama-3.1-Tulu-3-8B-Q8_0 - Text Generation 128
  CPU BLAS - Llama-3.1-Tulu-3-8B-Q8_0 - Prompt Processing 512
  CPU BLAS - Llama-3.1-Tulu-3-8B-Q8_0 - Prompt Processing 1024
  CPU BLAS - Llama-3.1-Tulu-3-8B-Q8_0 - Prompt Processing 2048
  CPU BLAS - Mistral-7B-Instruct-v0.3-Q8_0 - Text Generation 128
  CPU BLAS - Mistral-7B-Instruct-v0.3-Q8_0 - Prompt Processing 512
  CPU BLAS - Mistral-7B-Instruct-v0.3-Q8_0 - Prompt Processing 1024
  CPU BLAS - Mistral-7B-Instruct-v0.3-Q8_0 - Prompt Processing 2048
  CPU BLAS - granite-3.0-3b-a800m-instruct-Q8_0 - Text Generation 128
  CPU BLAS - granite-3.0-3b-a800m-instruct-Q8_0 - Prompt Processing 512
  CPU BLAS - granite-3.0-3b-a800m-instruct-Q8_0 - Prompt Processing 1024
  CPU BLAS - granite-3.0-3b-a800m-instruct-Q8_0 - Prompt Processing 2048