eoy2024

Benchmarks for a future article. AMD EPYC 4124P 4-Core testing with a Supermicro AS-3015A-I H13SAE-MF v1.00 (2.1 BIOS) and ASPEED on Ubuntu 24.04 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 2412201-NE-EOY20246700
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
Disable Color Branding
Prefer Vertical Bar Graphs

Additional Graphs

Show Perf Per Core/Thread Calculation Graphs Where Applicable
Show Perf Per Clock Calculation Graphs Where Applicable

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 05
  6 Hours, 48 Minutes
4484PX
December 07
  7 Hours, 3 Minutes
px
December 07
  7 Hours, 3 Minutes
4464p
December 09
  7 Hours, 21 Minutes
4464p epyc
December 09
  7 Hours, 21 Minutes
4364P
December 11
  8 Hours, 21 Minutes
4584PX
December 16
  6 Hours, 23 Minutes
EPYC 4584PX amd
December 17
  6 Hours, 23 Minutes
45
December 18
  6 Hours, 22 Minutes
41
December 19
  10 Hours, 3 Minutes
41 b
December 20
  13 Hours, 1 Minute
Invert Behavior (Only Show Selected Data)
  7 Hours, 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):


eoy2024ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerCompilerFile-SystemScreen Resolutiona4484PXpx4464p4464p epyc4364P4584PXEPYC 4584PX amd454141 bAMD EPYC 4564P 16-Core @ 5.88GHz (16 Cores / 32 Threads)Supermicro AS-3015A-I H13SAE-MF v1.00 (2.1 BIOS)AMD Device 14d82 x 32GB DRAM-4800MT/s Micron MTC20C2085S1EC48BA1 BC3201GB Micron_7450_MTFDKCC3T2TFS + 960GB SAMSUNG MZ1L2960HCJR-00A07ASPEEDAMD Rembrandt Radeon HD AudioVA24312 x Intel I210Ubuntu 24.046.8.0-11-generic (x86_64)GNOME Shell 45.3X Server 1.21.1.11GCC 13.2.0ext41024x768AMD EPYC 4484PX 12-Core @ 5.66GHz (12 Cores / 24 Threads)6.12.2-061202-generic (x86_64)AMD EPYC 4464P 12-Core @ 5.48GHz (12 Cores / 24 Threads)AMD EPYC 4364P 8-Core @ 5.57GHz (8 Cores / 16 Threads)AMD EPYC 4584PX 16-Core @ 5.76GHz (16 Cores / 32 Threads)AMD EPYC 4124P 4-Core @ 5.17GHz (4 Cores / 8 Threads)960GB SAMSUNG MZ1L2960HCJR-00A07 + 3201GB Micron_7450_MTFDKCC3T2TFSOpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --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 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-fxIygj/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-fxIygj/gcc-13-13.2.0/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-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- a: Scaling Governor: amd-pstate-epp performance (EPP: performance) - CPU Microcode: 0xa601209- 4484PX: Scaling Governor: amd-pstate-epp performance (Boost: Enabled EPP: performance) - CPU Microcode: 0xa601209- px: Scaling Governor: amd-pstate-epp performance (Boost: Enabled EPP: performance) - CPU Microcode: 0xa601209- 4464p: Scaling Governor: amd-pstate-epp performance (Boost: Enabled EPP: performance) - CPU Microcode: 0xa601209- 4464p epyc: Scaling Governor: amd-pstate-epp performance (Boost: Enabled EPP: performance) - CPU Microcode: 0xa601209- 4364P: Scaling Governor: amd-pstate-epp performance (Boost: Enabled EPP: performance) - CPU Microcode: 0xa601209- 4584PX: Scaling Governor: amd-pstate-epp performance (Boost: Enabled EPP: performance) - CPU Microcode: 0xa601209- EPYC 4584PX amd: Scaling Governor: amd-pstate-epp performance (Boost: Enabled EPP: performance) - CPU Microcode: 0xa601209- 45: Scaling Governor: amd-pstate-epp performance (Boost: Enabled EPP: performance) - CPU Microcode: 0xa601209- 41: Scaling Governor: amd-pstate-epp performance (Boost: Enabled EPP: performance) - CPU Microcode: 0xa601209- 41 b: Scaling Governor: amd-pstate-epp performance (Boost: Enabled EPP: performance) - CPU Microcode: 0xa601209Java Details- OpenJDK Runtime Environment (build 21.0.2+13-Ubuntu-2)Python Details- Python 3.12.3Security Details- a: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of Safe RET + 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 STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected- 4484PX: 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: Mitigation of Safe RET + 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; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected- px: 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: Mitigation of Safe RET + 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; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected- 4464p: 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: Mitigation of Safe RET + 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; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected- 4464p epyc: 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: Mitigation of Safe RET + 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; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected- 4364P: 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: Mitigation of Safe RET + 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; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected- 4584PX: 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: Mitigation of Safe RET + 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; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected- EPYC 4584PX amd: 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: Mitigation of Safe RET + 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; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected- 45: 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: Mitigation of Safe RET + 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; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected- 41: 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: Mitigation of Safe RET + 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; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected- 41 b: 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: Mitigation of Safe RET + 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; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected

a4484PXpx4464p4464p epyc4364P4584PXEPYC 4584PX amd454141 bResult OverviewPhoronix Test Suite100%213%326%439%RELIONACES DGEMMStockfishBYTE Unix BenchmarkApache CassandraBlenderOpenSSLASTC EncoderPrimesievePOV-RayEtcpakOSPRay7-Zip CompressionBuild2QuantLibNAMDSVT-AV1oneDNNGROMACSRustlsTimed Eigen CompilationCP2K Molecular Dynamicsx265Whisper.cppY-CruncherLiteRTXNNPACKONNX RuntimeApache CouchDBFinanceBenchNumpy BenchmarkRenaissanceGcrypt LibrarysimdjsonPyPerformance

eoy2024quantlib: Ssvt-av1: Preset 3 - Beauty 4K 10-bitrelion: Basic - CPUblender: Barbershop - CPU-Onlywhisper-cpp: ggml-medium.en - 2016 State of the Unioncp2k: H20-256couchdb: 500 - 3000 - 30quantlib: XXSwhisperfile: Mediumllamafile: wizardcoder-python-34b-v1.0.Q6_K - Prompt Processing 2048couchdb: 300 - 3000 - 30svt-av1: Preset 3 - Bosphorus 4Kbyte: Whetstone Doublewhisper-cpp: ggml-small.en - 2016 State of the Unionllamafile: wizardcoder-python-34b-v1.0.Q6_K - Text Generation 128blender: Pabellon Barcelona - CPU-Onlyllamafile: mistral-7b-instruct-v0.2.Q5_K_M - Prompt Processing 2048svt-av1: Preset 5 - Beauty 4K 10-bitcouchdb: 100 - 3000 - 30byte: Pipebyte: Dhrystone 2byte: System Callblender: Classroom - CPU-Onlyrustls: handshake-resume - TLS13_CHACHA20_POLY1305_SHA256rustls: handshake-ticket - TLS13_CHACHA20_POLY1305_SHA256llamafile: wizardcoder-python-34b-v1.0.Q6_K - 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 - Prompt Processing 2048xnnpack: QS8MobileNetV2xnnpack: FP16MobileNetV3Smallxnnpack: FP16MobileNetV3Largexnnpack: FP16MobileNetV2xnnpack: FP32MobileNetV3Smallxnnpack: FP32MobileNetV3Largexnnpack: FP32MobileNetV2xnnpack: FP32MobileNetV1xnnpack: FP16MobileNetV1rustls: handshake-ticket - TLS_ECDHE_ECDSA_WITH_AES_256_GCM_SHA384whisperfile: Smallrustls: handshake-resume - TLS_ECDHE_ECDSA_WITH_AES_256_GCM_SHA384openssl: ChaCha20openssl: ChaCha20-Poly1305openssl: AES-256-GCMopenssl: AES-128-GCMllamafile: mistral-7b-instruct-v0.2.Q5_K_M - Text Generation 128svt-av1: Preset 3 - Bosphorus 1080pcouchdb: 500 - 1000 - 30gcrypt: astcenc: Very Thoroughastcenc: Exhaustiveospray: particle_volume/scivis/real_timeospray: particle_volume/pathtracer/real_timesvt-av1: Preset 8 - Beauty 4K 10-bitgromacs: water_GMX50_barebuild2: Time To Compileprimesieve: 1e13llamafile: mistral-7b-instruct-v0.2.Q5_K_M - Prompt Processing 1024blender: Junkshop - CPU-Onlycassandra: Writesblender: Fishy Cat - CPU-Onlycouchdb: 300 - 1000 - 30pyperformance: async_tree_ioopenvino-genai: Gemma-7b-int4-ov - CPU - Time Per Output Tokenopenvino-genai: Gemma-7b-int4-ov - CPU - Time To First Tokenopenvino-genai: Gemma-7b-int4-ov - CPUllamafile: Llama-3.2-3B-Instruct.Q6_K - Prompt Processing 2048rustls: handshake - TLS_ECDHE_ECDSA_WITH_AES_256_GCM_SHA384llamafile: wizardcoder-python-34b-v1.0.Q6_K - Prompt Processing 512svt-av1: Preset 5 - Bosphorus 4Kospray: particle_volume/ao/real_timecp2k: Fayalite-FISTwhisper-cpp: ggml-base.en - 2016 State of the Unionpyperformance: asyncio_tcp_sslpyperformance: xml_etreenumpy: simdjson: Kostyallama-cpp: CPU BLAS - Llama-3.1-Tulu-3-8B-Q8_0 - Prompt Processing 1024llama-cpp: CPU BLAS - Mistral-7B-Instruct-v0.3-Q8_0 - Prompt Processing 1024blender: BMW27 - CPU-Onlyllama-cpp: CPU BLAS - Llama-3.1-Tulu-3-8B-Q8_0 - Text Generation 128llamafile: Llama-3.2-3B-Instruct.Q6_K - Text Generation 128namd: STMV with 1,066,628 Atomsllama-cpp: CPU BLAS - Mistral-7B-Instruct-v0.3-Q8_0 - Text Generation 128openvino-genai: Falcon-7b-instruct-int4-ov - CPU - Time Per Output Tokenopenvino-genai: Falcon-7b-instruct-int4-ov - CPU - Time To First Tokenopenvino-genai: Falcon-7b-instruct-int4-ov - CPUsvt-av1: Preset 13 - Beauty 4K 10-bitstockfish: Chess Benchmarkpyperformance: python_startupsimdjson: LargeRandmt-dgemm: Sustained Floating-Point Ratebuild-eigen: Time To Compilecouchdb: 100 - 1000 - 30onednn: Recurrent Neural Network Training - CPUrenaissance: ALS Movie Lensonednn: Recurrent Neural Network Inference - CPUsimdjson: DistinctUserIDstockfish: Chess Benchmarksimdjson: TopTweetsimdjson: PartialTweetsllamafile: TinyLlama-1.1B-Chat-v1.0.BF16 - Text Generation 128renaissance: In-Memory Database Shootoutrenaissance: Akka Unbalanced Cobwebbed Treerenaissance: Apache Spark PageRankllamafile: mistral-7b-instruct-v0.2.Q5_K_M - Prompt Processing 512cp2k: H20-64renaissance: Savina Reactors.IOrenaissance: Gaussian Mixture Modelpyperformance: gc_collectrenaissance: Apache Spark Bayesrenaissance: Finagle HTTP Requestsrenaissance: Rand Forestospray: gravity_spheres_volume/dim_512/scivis/real_timeospray: gravity_spheres_volume/dim_512/ao/real_timeonnx: ResNet101_DUC_HDC-12 - CPU - Standardonnx: ResNet101_DUC_HDC-12 - CPU - Standardrenaissance: Scala Dottyrenaissance: Genetic Algorithm Using Jenetics + Futuresonnx: fcn-resnet101-11 - CPU - Standardonnx: fcn-resnet101-11 - CPU - Standardonnx: GPT-2 - CPU - Standardonnx: GPT-2 - CPU - Standardospray: gravity_spheres_volume/dim_512/pathtracer/real_timeonnx: ZFNet-512 - CPU - Standardonnx: ZFNet-512 - CPU - Standardonnx: bertsquad-12 - CPU - Standardonnx: bertsquad-12 - CPU - Standardonnx: T5 Encoder - CPU - Standardonnx: T5 Encoder - CPU - Standardonnx: yolov4 - CPU - Standardonnx: yolov4 - CPU - Standardonnx: ArcFace ResNet-100 - CPU - Standardonnx: ArcFace ResNet-100 - CPU - Standardonnx: Faster R-CNN R-50-FPN-int8 - CPU - Standardonnx: Faster R-CNN R-50-FPN-int8 - CPU - Standardonnx: CaffeNet 12-int8 - CPU - Standardonnx: CaffeNet 12-int8 - CPU - Standardonnx: ResNet50 v1-12-int8 - CPU - Standardonnx: ResNet50 v1-12-int8 - CPU - Standardonnx: super-resolution-10 - CPU - Standardonnx: super-resolution-10 - CPU - Standardllama-cpp: CPU BLAS - granite-3.0-3b-a800m-instruct-Q8_0 - Prompt Processing 2048pyperformance: asyncio_websocketsllamafile: Llama-3.2-3B-Instruct.Q6_K - Prompt Processing 1024llamafile: wizardcoder-python-34b-v1.0.Q6_K - Prompt Processing 256llamafile: TinyLlama-1.1B-Chat-v1.0.BF16 - Prompt Processing 2048svt-av1: Preset 5 - Bosphorus 1080plitert: Inception V4litert: Inception ResNet V2litert: NASNet Mobilelitert: DeepLab V3litert: Mobilenet Floatlitert: SqueezeNetlitert: Quantized COCO SSD MobileNet v1litert: Mobilenet Quantrustls: handshake-ticket - TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256llama-cpp: CPU BLAS - Llama-3.1-Tulu-3-8B-Q8_0 - Prompt Processing 512llama-cpp: CPU BLAS - Mistral-7B-Instruct-v0.3-Q8_0 - Prompt Processing 512rustls: handshake-resume - TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256astcenc: Thoroughllamafile: wizardcoder-python-34b-v1.0.Q6_K - Text Generation 16financebench: Bonds OpenMPetcpak: Multi-Threaded - ETC2whisperfile: Tinyrustls: handshake - TLS13_CHACHA20_POLY1305_SHA256svt-av1: Preset 8 - Bosphorus 4Krustls: handshake - TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256pyperformance: django_templateopenvino-genai: Phi-3-mini-128k-instruct-int4-ov - CPU - Time Per Output Tokenopenvino-genai: Phi-3-mini-128k-instruct-int4-ov - CPU - Time To First Tokenopenvino-genai: Phi-3-mini-128k-instruct-int4-ov - CPUpovray: Trace Timenamd: ATPase with 327,506 Atomsllamafile: mistral-7b-instruct-v0.2.Q5_K_M - Prompt Processing 256pyperformance: raytracefinancebench: Repo OpenMPpyperformance: crypto_pyaespyperformance: chaospyperformance: gopyperformance: floatx265: Bosphorus 4Kpyperformance: regex_compilellamafile: Llama-3.2-3B-Instruct.Q6_K - Prompt Processing 512llama-cpp: CPU BLAS - granite-3.0-3b-a800m-instruct-Q8_0 - Prompt Processing 1024y-cruncher: 1Bpyperformance: pathlibpyperformance: pickle_pure_pythononednn: Deconvolution Batch shapes_1d - CPUllamafile: mistral-7b-instruct-v0.2.Q5_K_M - Text Generation 16llamafile: TinyLlama-1.1B-Chat-v1.0.BF16 - Prompt Processing 1024pyperformance: json_loadsastcenc: Fastpyperformance: nbodycompress-7zip: Decompression Ratingcompress-7zip: Compression Ratingsvt-av1: Preset 13 - Bosphorus 4Ksvt-av1: Preset 8 - Bosphorus 1080ponednn: IP Shapes 1D - CPUastcenc: Mediumllama-cpp: CPU BLAS - granite-3.0-3b-a800m-instruct-Q8_0 - Text Generation 128llamafile: Llama-3.2-3B-Instruct.Q6_K - Prompt Processing 256llama-cpp: CPU BLAS - granite-3.0-3b-a800m-instruct-Q8_0 - Prompt Processing 512llamafile: Llama-3.2-3B-Instruct.Q6_K - Text Generation 16y-cruncher: 500Mprimesieve: 1e12llamafile: TinyLlama-1.1B-Chat-v1.0.BF16 - Prompt Processing 512llamafile: TinyLlama-1.1B-Chat-v1.0.BF16 - Text Generation 16onednn: IP Shapes 3D - CPUx265: Bosphorus 1080ponednn: Convolution Batch Shapes Auto - CPUsvt-av1: Preset 13 - Bosphorus 1080pllamafile: TinyLlama-1.1B-Chat-v1.0.BF16 - Prompt Processing 256onednn: Deconvolution Batch shapes_3d - CPUopenssl: SHA256a4484PXpx4464p4464p epyc4364P4584PXEPYC 4584PX amd454141 b12.74761.422944.27506.2700.91592.857511.77513.432534.91912288367.839.59343491.9245.078381.99166.12327686.504232.18848806257.11866536062.749140426.6143.36388077.69404263.45614463.0962.978449201498119097918101495125211431553632.14195.416421820810.21130588495050923935293409717275170010478452217010.4729.573148.049162.1252.7411.68448.98486236.24512.4681.69292.05378.4981638473.5627133371.35106.13755101.72106.629.8332768423535.68307234.5389.0091794.03287.4897364535.8775.755.9770.8569.2653.556.8820.130.756567.2477.3486.0612.9318.588547527965.771.831141.19410458.65569.9291372.039805.7700.85910.464650703810.469.7626.283256.14403.82412.2819258.1913506.43399.5677490.02319.4414.47.587897.63944648.5221.54196477.0732.8310.8753.21677.42776134.5968.820939.76985102.33164.14115.58996.39112156.45390.452311.055223.55342.453721.242947.06911.57084636.3182.55898390.5977.08601141.117279.0431516384153632768101.97121477.819530.2169363579.671211.481794.112129.52823.17262033270.7668.43563852.5720.30251.7833061.21875577.81741.7093576454.45102.00580462.620.751.8655.9319.2818.5422.79632409617521418.44531241.738.277.850.732.5769.88192355.0918.48514.21652.9761210.221638412.1396.649559165916163859212.52339.0231.12573156.221747.724096327.319.038.7726.347819224.594.058114.456.67287842.55840962.4129411.86471.188729.4679.34809.78969628.104559.34612.1169473.5509112288406.127.684244075.3268.238912.05226.34327685.602253.9933443359.21346521770.330761218.9197.2333882.92344296.24614463.863.617177791467121780915151365125713831329363.1173.381971586292.429710523569068816544020711602918707649633676010.9125.446164.468171.0231.94121.18876.44913199.02310.9671.577111.651110.6081638497.0117496096.67117.56666697.79121.4810.2332768306153.2307229.0946.5277692.2192.7093359036.8745.596.1166.5766.8574.087.1120.390.651197.4174.6593.0113.417.406452675466.081.84842.73064267.36475.9011898.369378.8965.01510.763370229810.8210.127.593241.54038.42138.1819253.0053655.83860.6699513.22492.2422.05.548885.63122850.1411.17627428.6904.0355.7512.810936.25815159.716.411989.01322110.9468.905114.51224.80287208.17493.160510.733826.747837.383224.940240.09351.06188941.4012.80544356.4097.98873125.172222.753211638415363276888.41522083.319477.88057.562343.381244.71809.181420.15848.9432282729.6469.1168.23035330.2114.171.8334600.773438410.72637.1346257716.6485.20159308.752149.3158.9120.2825.2642.38124409618222320.33203143.139.778.651.327.1671.78192232.2618.37914.41693.4029310.451638412.4278.244559.5125698141263198.112287.0471.93806109.026552.34096243.1419.498.6889.116819225.862.73072101.374.11551776.11540963.508411.8391.184733.02678.4809.489631.31560.712.1057475.5108412288408.4837.646244131266.814252.05224.64327685.551254.73333381363.11340340196.630701622.8197.53333574.3342775.29614463.7963.417237981527124883715741368127213861340712.85167.892191572010.689701989745068678955550709026564807618440561010.9325.447164.812163.8391.93911.18626.52304197.210.8551.575113.78110.7091638497.117394697.09119.34965697.61122.310.2432768304060.28307228.8246.5220694.89693.4546359036.5831.425.4566.3566.5273.167.1220.510.654487.4474.549313.4117.355429733966.091.84842.01283167.07676.3891895.689275.7966.0138.973387159510.518.3527.83175.64002.32229.7819252.7243676.03815.2706474.92483.1453.25.61475.71084854.3341.1705436.2920.7357.6022.796386.33034157.8936.40749.01687110.89268.610414.57474.85142206.09193.344110.712726.948537.104823.060443.3621.066937.7782.80695356.1947.99486125.076208.993221638415363276888.2722752.419490.77931.642359.991244.511821.351417.35849.2092292879.4467.9568.813038723.4814.14641.8434896.835938409.87538.7182857688.0884.99859206.3421.249.2858.8620.2925.3282.35379409618222318.73828143.339.479.450.826.9472.58192244.7718.36514.41683.4062810.451638412.5277.299459.2125605142213194.024286.9621.93913108.858852.374096232.8619.58.6239.147819225.942.72942101.254.13321769.81840963.5124310.28731.129937.33739.91838.18425676.742551.34610.5788491.9841212288401.9897.252242679280.034192.05243.35327685.445259.19632471450.7127293878829969810.9212.3328877.92336267.09614460.6860.477037591429120879914731357126714181297646.72178.329171543373.129395674371066548747470674315359607270995422010.9324.854164.358156.3431.85461.13656.29615199.51210.9141.398121.737118.66416384104.7173633105.45118.16173898.33124.3810.1732768301017.23307227.366.28807100.48697.9440862635.3781.665.7863.7164.1579.197.0920.480.614527.4274.949313.3418.002431588535.81.81818.67814169.37976.6961911.449243.01009.5810.543031597210.669.827.363288.54084.62366.1819258.3973523.13647.2671515.32650.9415.35.357565.47043872.7421.14581424.6924.5386.5632.586887.33331136.2996.173999.85896101.41368.077514.68855.95035168.03695.250710.498227.682836.120521.570846.35431.3797724.5332.77422360.3547.41037134.941213.023151638415363276886.9623141.420123.57759.692272.281248.121820.341431.19859.7572283083.5464.9665.243061522.2613.5231.8333551.523438388.40440.84756047.3882.52557954.1920.650.4559.8719.8226.6092.21645409617721588.9062541.83978.349.726.0969.58192235.0119.36414.11643.4328710.481638412263.547557.2118569135033192.171282.0081.88731103.808950.54096235.9419.479.0419.701819225.773.75003101.236.36475752.15540963.3895810.25311.132946.43731.39839.98575678.676551.610.5616492.6951912288405.4487.232242340.5280.472062.06245.12327685.422259.05532439611.61263644317.829923094.4211.4326342.42335411.96614460.860.37037661427121079115421327124614141300937.3177.871532788.049390757287066497739430673962543807266752730010.9124.805163.363156.7121.85591.13686.27188199.54610.8851.393117.067118.84816384104.01174025105.24116.41774298.03124.5210.232768301217.61307227.5646.2921197.44697.5456262835.6783.436.0263.562.8479.247.120.50.610937.4475.0694.7713.3217.944420429515.811.73812.14708969.57876.5771913.868980.81012.1910.282994007410.699.6327.353275.54098.52373.0819258.8643688.63703.2671513.32708.1412.95.361895.47057864.811.15632442.9884.0386.0612.590257.36544135.7186.176249.83893101.61868.093414.6856.0047166.51295.357610.486527.472936.396821.693346.09211.33516748.662.76714361.2977.43372134.517212.33161638415363276886.38622928.320503.57640.812289.581248.61822.71415.11857.4892304145.7964.4364.683082657.513.51391.8333538.269531388.12838.5784356176.982.16458056.0120.650.4659.8619.8226.5952.20728409617621612.0429694238.67649.725.85708192227.7119.27814.11643.4254710.451638412263.366258.6118587133804190.343281.8921.88919103.674450.434096242.119.539.0679.705819225.783.74578101.336.3715748.14740963.388238.026960.8721369.225978.03966.00638740.089545.9678.19153528.3326212288398.3865.373174760.6314.477221.93320.59327683.802239.07823448988.5966576789.421559197276.44272035.68265949.2614455.9556.0540045510671088335753708105615781187839.34193.936641298373.666804832966048278109930499309456005375566370010.3218.744163.915154.3171.40520.86124.62423177.1186.4581.104154.429150.66816384141.99139110136.51118.317726105.09154.839.5232768225379.66307219.854.63965108.436103.7154961034.7813.316.0659.2759.54103.976.6419.60.497717.0279.41115.7312.599.188315149376.341.81629.1651284.47773.2372422.835993.41270.7610.852401702410.8610.0425.523244.83720.12506.8819268.1713349.53397.0649483.41705.5381.63.831523.917491091.320.916317464.5749.9487.3012.052117.96069125.524.5231810.283597.226850.469819.81277.06075141.56875.541413.237225.584539.082822.319244.80021.31121762.2482.65482376.5316.36564157.078171.683081638415363276865.31824502.121153.53808.81569.751044.531654.651494.51935.9942353677.9660.8761.092874011.5610.33921.7632715.808594304.24441.2956243571.8665.19445095.3520.152.6471.581934.9171.74038409617421121.61328140.738.375.348.218.8568.18192185.9723.88613.51623.786299.921638411.8205.014456.888229101872156.598215.2131.9068680.368651.864096204.0618.8310.95412.273819224.274.4509688.717.87367652.96540963.7800814.01711.426551.722514.7686.87944517.664553.80514.624425.5096912288397.4269.329329601.4233.684412.03171.82327686.327252.82544169926.71765175324.640531084149.37379874.32393408.85614486.1786.568098771527122493317601484119412001559859.89155.139231863395.07127793294810907114471409316808759010045000881010.8129.067163.805150.6392.56341.56978.49097218.18112.0861.81791.72484.4891638473.8723589373.96116.24165398.91105.3610.1132768435194.2307234.1648.5157384.80284.6839858236.1806.836.0790.0191.1255.26720.230.759877.3275.0686.0413.3217.845572477636.011.821091.01150958.81875.8931320.999891.4672.50611.054444645811.259.8527.23232.84384.02182.5819243.1673582.43806.5682477.02627.3420.77.323537.3876650.961.53619441.91050.3295.2813.386586.02069166.0478.437528.87444112.67166.617215.01064.90285203.94788.063211.355122.806243.843620.718348.26360.9941931005.52.60974383.1348.18519122.168289.413151638415363276898.78920102.8182249311.772185.781165.51769.561337.78734.3522528614.5292.8993.243397871.0318.70751.8234053.0625556.04335.9149972793.26101.33274468.4920.850.8456.0619.6719.552.83913409617621948.93554742.739.278.451.132.5971.48192290.117.34314.11663.0721910.31638412.5368.545159164581170361217.908320.1841.12465144.478249.94096316.819.278.1736.926819225.322.74699113.374.03808842.34440962.6721914.04721.433553.782513.66684.38662517.918554.39714.6973424.9466612288400.3999.4331202233.053162.03171.25327686.391253.493443239201771958344.740718581.6148.97382691.63394570614486.4786.448078711542122293117561477119211961569308.65155.97451874050.7128265416980909274180309355599135010066659164010.8529.205161.089159.2472.56941.57438.51233220.79911.8971.80892.02184.2631638473.5523822773.86116.61765198.64104.3810.1432768436101.34307234.4148.5714684.21184.0281358636.2754.416.0491.3190.4855.74720.260.756457.3775.5186.213.2417.863586766366.011.831093.09786758.69276.6691320.379969.8670.23411.14457247949.199.9827.193442.94344.82227.9819242.513641.93773.1686515.62571.2402.27.322847.44575647.8141.54365411.9959.3294.9273.390646.03024165.7698.477278.82581113.29466.920214.94274.91381203.49188.362211.316622.698444.05120.881647.8850.999701999.9732.61221382.7718.18265122.205289.043171638415363276899.44519859.418412.39231.972200.931175.51765.391339.69742.0292558621.3893.9193.493431395.7318.74991.8333913.464844556.41235.426272508.13101.53774683.742150.6855.4719.7320.3252.85485409617621848.07031242.739.179.65132.2269.88192324.0417.56714.11653.0634110.371638412.2368.573458.8165934169750218.529327.2641.12801144.626249.764096326.1619.248.2796.903819225.32.77727113.164.04877845.58440962.6642914.04311.427549.8513.54684.13056517.502552.20914.6226424.5745912288399.2619.345331193.7232.583532.03170.09327686.322253.27944309550.91766450267.140711986.9149.37382067.3395112.39614486.6886.918018981547122392917541478118911951569479.7154.436581862938128278766110910250928809361459418010074230129010.8229.179161.473167.6312.56961.57678.54395220.72812.1121.81191.01784.3151638473.6923682673.48116.22263398.31104.8710.1732768435926.4307234.668.5487884.07284.1127658136.1833.645.5490.9691.1755.18720.290.762277.3774.8687.3413.3617.836557627026.031.831092.08849557.1876.6621320.239972.0670.79.15487970799.410.3227.213408.44359.32179.0819242.4693589.43829.9689500.02570.5407.77.315947.46399645.2081.54988455.6891.2293.9613.401786.01631166.1688.496468.83524113.17167.184914.88374.90313203.93488.410411.310522.992343.488620.975347.67210.9866431013.192.62573380.8018.18175122.219273.673141638415363276898.9051968617824.69245.762152.481178.941763.221343.22730.1672559245.291.7194.083379343.7218.76611.8633906.6875556.86535.1103672140.75101.52274672.5720.950.7455.6719.7119.5212.83798409617922060.15234442.438.878.150.732.2370.48192291.0217.36414.21633.0591810.341638412.3369.027459.6166180169373217.957325.9451.13012144.613150.334096337.0619.168.1196.937819225.382.77835113.314.0355835.4540962.662964.430210.4432771.7782077.691281.082695.3524.47643515.4952.80884017.4581.27325667.712.178314.48511669425.7478992410.710707715560.89135362.96132265.65623714201821314121342142523673512596141.19732992.323328083586023701092760238456531602537612329010.578212.829165.5470.69230.42442.29544112.724.3340.664300.72301.596292.1466243280.26153.029855106833.7810.8692.30223177.144187.2864572336.7705.995.7215.530.273528.007126302646.931.7259.146491145.2196.7444726.087465.12415.259.921202468210.029.322951.14474.82267.1123.6563745.83087.0702566.21664.4439.91.908041.844072151.820.464723506.5806.4926.7561.079038.04835124.1662.2385113.532373.884595.066410.51867.04122141.955139.6597.1601541.941823.841227.689136.1072.01609495.8364.8153207.60513.645773.270733037.17149048.8418445153.492950.82193.883173.552736.981781.61656876.392105346.655.065547198.941406155.44121499.7335.84922255.2321.270.0640.8889518232271.6660164339.978.452.810.9172.640.56214.61717.2673812.4100.365960.4437815407985.948123.7923.8311739.320518.16524.7234.9468956.619.18658380.6337.536554.430910.4413046.8052095.981797.86251282.66698.0744.46411845.1712288513.5832.79383540.7583.20451.93671.42327682.172316.45911518844.2475486444.110629944.8563.14135016.23132025.36614430.9331.04623720199820974141359144324073508602002.02310.38181690282.023328954286023701138360237605350102558237464010.4610.554212.528169.1410.69060.42232.27718111.4834.3320.661294.951302.73716384294.0466519282.19154.041858124.24288.648.0532768105324.92307210.8052.26612179.122186.2290673736.6700.765.6831.9132.59214.896.5520.080.273206.987.9217.8511.387.984149437376.971.71263.204392146.65596.5544752.857147.42441.2710.16116075029.979.325.093681.94457.12292.28192130.8133818.53130.4704600.11669.9485.41.894691.904812152.920.464485515.8801.4927.3881.078298.10767123.2692.2231813.695573.004794.121610.62417.13857140.034139.7187.1571241.819923.910825.949938.52942.01913495.0784.94636202.11813.541773.8364105.13311638415363276837.0749101.941757.95034.382994.592190.793179.552741.681785.41571309.6232.9732.942103587.55.04481.7346851.421875154.24866.1132321444.0935.64122201.0321.360.07130.4716.6570.7740.89241409618032477.34960943.140.278.752.910.8972.78192115.6739.24514.61717.281399.991638412.499.399960432105342585.603122.9213.8288539.037751.54096122.819.1718.2924.821819223.84.9586756.039.16956377.79140967.55315OpenBenchmarking.org

QuantLib

OpenBenchmarking.orgtasks/s, More Is BetterQuantLib 1.35-devSize: SpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b414812162011.8390012.7476014.0472014.0171014.0431011.8647010.2531010.287308.026964.430914.430211. (CXX) g++ options: -O3 -march=native -fPIE -pie

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.3Encoder Mode: Preset 3 - Input: Beauty 4K 10-bitpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b410.32240.64480.96721.28961.6121.1841.4221.4331.4261.4271.1881.1321.1290.8720.4410.4431. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

RELION

OpenBenchmarking.orgSeconds, Fewer Is BetterRELION 5.0Test: Basic - Device: CPUpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b417001400210028003500733.02944.27553.78551.72549.80729.40946.43937.331369.233046.812771.781. (CXX) g++ options: -fPIC -std=c++14 -fopenmp -O3 -rdynamic -lfftw3f -lfftw3 -ldl -ltiff -lpng -ljpeg -lmpi_cxx -lmpi

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.3Blend File: Barbershop - Compute: CPU-OnlypxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41400800120016002000678.40506.20513.66514.70513.54679.34731.39739.91978.032095.982077.69

Whisper.cpp

OpenBenchmarking.orgSeconds, Fewer Is BetterWhisper.cpp 1.6.2Model: ggml-medium.en - Input: 2016 State of the UnionpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b400800120016002000809.49700.91684.39686.88684.13809.79839.99838.18966.011797.861. (CXX) g++ options: -O3 -std=c++11 -fPIC -pthread -msse3 -mssse3 -mavx -mf16c -mfma -mavx2 -mavx512f -mavx512cd -mavx512vl -mavx512dq -mavx512bw -mavx512vbmi -mavx512vnni

CP2K Molecular Dynamics

OpenBenchmarking.orgSeconds, Fewer Is BetterCP2K Molecular Dynamics 2024.3Input: H20-256pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4130060090012001500631.31592.86517.92517.66517.50628.10678.68676.74740.091282.661281.081. (F9X) gfortran options: -fopenmp -march=native -mtune=native -O3 -funroll-loops -fbacktrace -ffree-form -fimplicit-none -std=f2008 -lcp2kstart -lcp2kmc -lcp2kswarm -lcp2kmotion -lcp2kthermostat -lcp2kemd -lcp2ktmc -lcp2kmain -lcp2kdbt -lcp2ktas -lcp2kgrid -lcp2kgriddgemm -lcp2kgridcpu -lcp2kgridref -lcp2kgridcommon -ldbcsrarnoldi -ldbcsrx -lcp2kdbx -lcp2kdbm -lcp2kshg_int -lcp2keri_mme -lcp2kminimax -lcp2khfxbase -lcp2ksubsys -lcp2kxc -lcp2kao -lcp2kpw_env -lcp2kinput -lcp2kpw -lcp2kgpu -lcp2kfft -lcp2kfpga -lcp2kfm -lcp2kcommon -lcp2koffload -lcp2kmpiwrap -lcp2kbase -ldbcsr -lsirius -lspla -lspfft -lsymspg -lvdwxc -l:libhdf5_fortran.a -l:libhdf5.a -lz -lgsl -lelpa_openmp -lcosma -lcosta -lscalapack -lxsmmf -lxsmm -ldl -lpthread -llibgrpp -lxcf03 -lxc -lint2 -lfftw3_mpi -lfftw3 -lfftw3_omp -lmpi_cxx -lmpi -l:libopenblas.a -lvori -lstdc++ -lmpi_usempif08 -lmpi_mpifh -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm

Apache CouchDB

OpenBenchmarking.orgSeconds, Fewer Is BetterApache CouchDB 3.4.1Bulk Size: 500 - Inserts: 3000 - Rounds: 30pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41150300450600750560.70511.78554.40553.81552.21559.35551.60551.35545.97698.07695.351. (CXX) g++ options: -flto -lstdc++ -shared -lei

QuantLib

OpenBenchmarking.orgtasks/s, More Is BetterQuantLib 1.35-devSize: XXSpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b414812162012.1057013.4320014.6973014.6240014.6226012.1169010.5616010.578808.191534.464114.476431. (CXX) g++ options: -O3 -march=native -fPIE -pie

Whisperfile

OpenBenchmarking.orgSeconds, Fewer Is BetterWhisperfile 20Aug24Model Size: MediumpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b2004006008001000475.51534.92424.95425.51424.57473.55492.70491.98528.33845.17

Llamafile

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.8.16Model: wizardcoder-python-34b-v1.0.Q6_K - Test: Prompt Processing 2048pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b3K6K9K12K15K12288122881228812288122881228812288122881228812288

Apache CouchDB

OpenBenchmarking.orgSeconds, Fewer Is BetterApache CouchDB 3.4.1Bulk Size: 300 - Inserts: 3000 - Rounds: 30pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41110220330440550408.48367.83400.40397.43399.26406.12405.45401.99398.39513.58515.501. (CXX) g++ options: -flto -lstdc++ -shared -lei

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.3Encoder Mode: Preset 3 - Input: Bosphorus 4KpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4136912157.6469.5909.4009.3299.3457.6847.2327.2525.3732.7932.8081. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

BYTE Unix Benchmark

OpenBenchmarking.orgMWIPS, More Is BetterBYTE Unix Benchmark 5.1.3-gitComputational Test: Whetstone DoublepxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4170K140K210K280K350K244131.0343491.9331202.0329601.4331193.7244075.3242340.5242679.0174760.683540.784017.41. (CC) gcc options: -pedantic -O3 -ffast-math -march=native -mtune=native -lm

Whisper.cpp

OpenBenchmarking.orgSeconds, Fewer Is BetterWhisper.cpp 1.6.2Model: ggml-small.en - Input: 2016 State of the UnionpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41130260390520650266.81245.08233.05233.68232.58268.24280.47280.03314.48583.20581.271. (CXX) g++ options: -O3 -std=c++11 -fPIC -pthread -msse3 -mssse3 -mavx -mf16c -mfma -mavx2 -mavx512f -mavx512cd -mavx512vl -mavx512dq -mavx512bw -mavx512vbmi -mavx512vnni

Llamafile

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.8.16Model: wizardcoder-python-34b-v1.0.Q6_K - Test: Text Generation 128pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b0.46350.9271.39051.8542.31752.051.992.032.032.032.052.062.051.931.93

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.3Blend File: Pabellon Barcelona - Compute: CPU-OnlypxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41140280420560700224.64166.12171.25171.82170.09226.34245.12243.35320.59671.42667.71

Llamafile

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.8.16Model: mistral-7b-instruct-v0.2.Q5_K_M - Test: Prompt Processing 2048pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b7K14K21K28K35K32768327683276832768327683276832768327683276832768

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.3Encoder Mode: Preset 5 - Input: Beauty 4K 10-bitpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b412468105.5516.5046.3916.3276.3225.6025.4225.4453.8022.1722.1781. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Apache CouchDB

OpenBenchmarking.orgSeconds, Fewer Is BetterApache CouchDB 3.4.1Bulk Size: 100 - Inserts: 3000 - Rounds: 30pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4170140210280350254.73232.19253.49252.83253.28253.99259.06259.20239.08316.46314.491. (CXX) g++ options: -flto -lstdc++ -shared -lei

BYTE Unix Benchmark

OpenBenchmarking.orgLPS, More Is BetterBYTE Unix Benchmark 5.1.3-gitComputational Test: PipepxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4110M20M30M40M50M33381363.148806257.144323920.044169926.744309550.933443359.232439611.632471450.723448988.511518844.211669425.71. (CC) gcc options: -pedantic -O3 -ffast-math -march=native -mtune=native -lm

OpenBenchmarking.orgLPS, More Is BetterBYTE Unix Benchmark 5.1.3-gitComputational Test: Dhrystone 2pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41400M800M1200M1600M2000M1340340196.61866536062.71771958344.71765175324.61766450267.11346521770.31263644317.81272938788.0966576789.4475486444.1478992410.71. (CC) gcc options: -pedantic -O3 -ffast-math -march=native -mtune=native -lm

OpenBenchmarking.orgLPS, More Is BetterBYTE Unix Benchmark 5.1.3-gitComputational Test: System CallpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4111M22M33M44M55M30701622.849140426.640718581.640531084.040711986.930761218.929923094.429969810.921559197.010629944.810707715.01. (CC) gcc options: -pedantic -O3 -ffast-math -march=native -mtune=native -lm

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.3Blend File: Classroom - Compute: CPU-OnlypxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41120240360480600197.53143.36148.97149.37149.37197.20211.40212.30276.44563.14560.89

Rustls

OpenBenchmarking.orghandshakes/s, More Is BetterRustls 0.23.17Benchmark: handshake-resume - Suite: TLS13_CHACHA20_POLY1305_SHA256pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4180K160K240K320K400K333574.30388077.69382691.63379874.32382067.30333882.92326342.42328877.92272035.68135016.23135362.961. (CC) gcc options: -m64 -lgcc_s -lutil -lrt -lpthread -lm -ldl -lc -pie -nodefaultlibs

OpenBenchmarking.orghandshakes/s, More Is BetterRustls 0.23.17Benchmark: handshake-ticket - Suite: TLS13_CHACHA20_POLY1305_SHA256pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4190K180K270K360K450K342775.29404263.45394570.00393408.85395112.39344296.24335411.96336267.09265949.20132025.36132265.651. (CC) gcc options: -m64 -lgcc_s -lutil -lrt -lpthread -lm -ldl -lc -pie -nodefaultlibs

Llamafile

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.8.16Model: wizardcoder-python-34b-v1.0.Q6_K - Test: Prompt Processing 1024pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b130026003900520065006144614461446144614461446144614461446144

Llama.cpp

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b4154Backend: CPU BLAS - Model: Llama-3.1-Tulu-3-8B-Q8_0 - Test: Prompt Processing 2048pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b2040608010063.7963.0986.4786.1786.6863.8060.8060.6855.9530.931. (CXX) g++ options: -std=c++11 -fPIC -O3 -pthread -fopenmp -march=native -mtune=native -lopenblas

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b4154Backend: CPU BLAS - Model: Mistral-7B-Instruct-v0.3-Q8_0 - Test: Prompt Processing 2048pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b2040608010063.4162.9786.4486.5686.9163.6160.3060.4756.0531.041. (CXX) g++ options: -std=c++11 -fPIC -O3 -pthread -fopenmp -march=native -mtune=native -lopenblas

XNNPACK

OpenBenchmarking.orgus, Fewer Is BetterXNNPACK b7b048Model: QS8MobileNetV2pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4120040060080010007238448078148017177037034006236231. (CXX) g++ options: -O3 -lrt -lm

OpenBenchmarking.orgus, Fewer Is BetterXNNPACK b7b048Model: FP16MobileNetV3SmallpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4120040060080010007989208718828987797667594557207141. (CXX) g++ options: -O3 -lrt -lm

OpenBenchmarking.orgus, Fewer Is BetterXNNPACK b7b048Model: FP16MobileNetV3LargepxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41400800120016002000152714981542153215471467142714291067199820181. (CXX) g++ options: -O3 -lrt -lm

OpenBenchmarking.orgus, Fewer Is BetterXNNPACK b7b048Model: FP16MobileNetV2pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b415001000150020002500124811901222122412231217121012081088209721311. (CXX) g++ options: -O3 -lrt -lm

OpenBenchmarking.orgus, Fewer Is BetterXNNPACK b7b048Model: FP32MobileNetV3SmallpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4120040060080010008379799319349298097917993354144121. (CXX) g++ options: -O3 -lrt -lm

OpenBenchmarking.orgus, Fewer Is BetterXNNPACK b7b048Model: FP32MobileNetV3LargepxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4140080012001600200015741810175617601754151515421473753135913421. (CXX) g++ options: -O3 -lrt -lm

OpenBenchmarking.orgus, Fewer Is BetterXNNPACK b7b048Model: FP32MobileNetV2pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b413006009001200150013681495147714841478136513271357708144314251. (CXX) g++ options: -O3 -lrt -lm

OpenBenchmarking.orgus, Fewer Is BetterXNNPACK b7b048Model: FP32MobileNetV1pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b415001000150020002500127212521192119411891257124612671056240723671. (CXX) g++ options: -O3 -lrt -lm

OpenBenchmarking.orgus, Fewer Is BetterXNNPACK b7b048Model: FP16MobileNetV1pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b418001600240032004000138611431196120011951383141414181578350835121. (CXX) g++ options: -O3 -lrt -lm

Rustls

OpenBenchmarking.orghandshakes/s, More Is BetterRustls 0.23.17Benchmark: handshake-ticket - Suite: TLS_ECDHE_ECDSA_WITH_AES_256_GCM_SHA384pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41300K600K900K1200K1500K1340712.851553632.141569308.651559859.891569479.701329363.101300937.301297646.721187839.34602002.02596141.191. (CC) gcc options: -m64 -lgcc_s -lutil -lrt -lpthread -lm -ldl -lc -pie -nodefaultlibs

Whisperfile

OpenBenchmarking.orgSeconds, Fewer Is BetterWhisperfile 20Aug24Model Size: SmallpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b70140210280350167.89195.42155.97155.14154.44173.38177.87178.33193.94310.38

Rustls

OpenBenchmarking.orghandshakes/s, More Is BetterRustls 0.23.17Benchmark: handshake-resume - Suite: TLS_ECDHE_ECDSA_WITH_AES_256_GCM_SHA384pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41400K800K1200K1600K2000K1572010.681820810.211874050.701863395.071862938.001586292.421532788.041543373.121298373.66690282.02732992.321. (CC) gcc options: -m64 -lgcc_s -lutil -lrt -lpthread -lm -ldl -lc -pie -nodefaultlibs

OpenSSL

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSLAlgorithm: ChaCha20pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4130000M60000M90000M120000M150000M970198974501305884950501282654169801277932948101282787661109710523569093907572870939567437106804832966033289542860332808358601. OpenSSL 3.0.13 30 Jan 2024 (Library: OpenSSL 3.0.13 30 Jan 2024) - Additional Parameters: -engine qatengine -async_jobs 8

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSLAlgorithm: ChaCha20-Poly1305pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4120000M40000M60000M80000M100000M68678955550923935293409092741803090711447140910250928806881654402066497739430665487474704827810993023701138360237010927601. OpenSSL 3.0.13 30 Jan 2024 (Library: OpenSSL 3.0.13 30 Jan 2024) - Additional Parameters: -engine qatengine -async_jobs 8

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSLAlgorithm: AES-256-GCMpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4120000M40000M60000M80000M100000M70902656480971727517009355599135093168087590936145941807116029187067396254380674315359604993094560023760535010238456531601. OpenSSL 3.0.13 30 Jan 2024 (Library: OpenSSL 3.0.13 30 Jan 2024) - Additional Parameters: -engine qatengine -async_jobs 8

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSLAlgorithm: AES-128-GCMpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4120000M40000M60000M80000M100000M761844056101047845221701006665916401004500088101007423012907649633676072667527300727099542205375566370025582374640253761232901. OpenSSL 3.0.13 30 Jan 2024 (Library: OpenSSL 3.0.13 30 Jan 2024) - Additional Parameters: -engine qatengine -async_jobs 8

Llamafile

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.8.16Model: mistral-7b-instruct-v0.2.Q5_K_M - Test: Text Generation 128pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b369121510.9310.4710.8510.8110.8210.9110.9110.9310.3210.46

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.3Encoder Mode: Preset 3 - Input: Bosphorus 1080ppxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4171421283525.4529.5729.2129.0729.1825.4524.8124.8518.7410.5510.581. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Apache CouchDB

OpenBenchmarking.orgSeconds, Fewer Is BetterApache CouchDB 3.4.1Bulk Size: 500 - Inserts: 1000 - Rounds: 30pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4150100150200250164.81148.05161.09163.81161.47164.47163.36164.36163.92212.53212.831. (CXX) g++ options: -flto -lstdc++ -shared -lei

Gcrypt Library

OpenBenchmarking.orgSeconds, Fewer Is BetterGcrypt Library 1.10.3pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b414080120160200163.84162.13159.25150.64167.63171.02156.71156.34154.32169.14165.551. (CC) gcc options: -O2 -fvisibility=hidden

ASTC Encoder

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 5.0Preset: Very ThoroughpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b410.61671.23341.85012.46683.08351.93912.74102.56942.56342.56961.94121.85591.85461.40520.69060.69231. (CXX) g++ options: -O3 -flto -pthread

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 5.0Preset: ExhaustivepxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b410.3790.7581.1371.5161.8951.18621.68441.57431.56971.57671.18871.13681.13650.86120.42230.42441. (CXX) g++ options: -O3 -flto -pthread

OSPRay

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 3.2Benchmark: particle_volume/scivis/real_timepxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4136912156.523048.984868.512338.490978.543956.449136.271886.296154.624232.277182.29544

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 3.2Benchmark: particle_volume/pathtracer/real_timepxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4150100150200250197.20236.25220.80218.18220.73199.02199.55199.51177.12111.48112.72

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.3Encoder Mode: Preset 8 - Input: Beauty 4K 10-bitpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41369121510.85512.46811.89712.08612.11210.96710.88510.9146.4584.3324.3341. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

GROMACS

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACSInput: water_GMX50_barepxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b410.40880.81761.22641.63522.0441.5751.6921.8081.8171.8111.5771.3931.3981.1040.6610.6641. GROMACS version: 2023.3-Ubuntu_2023.3_1ubuntu3

Build2

OpenBenchmarking.orgSeconds, Fewer Is BetterBuild2 0.17Time To CompilepxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4170140210280350113.7892.0592.0291.7291.02111.65117.07121.74154.43294.95300.72

Primesieve

OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 12.6Length: 1e13pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4170140210280350110.7178.5084.2684.4984.32110.61118.85118.66150.67302.74301.601. (CXX) g++ options: -O3

Llamafile

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.8.16Model: mistral-7b-instruct-v0.2.Q5_K_M - Test: Prompt Processing 1024pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4K8K12K16K20K16384163841638416384163841638416384163841638416384

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.3Blend File: Junkshop - Compute: CPU-OnlypxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b416012018024030097.1073.5673.5573.8773.6997.01104.01104.70141.99294.04292.14

Apache Cassandra

OpenBenchmarking.orgOp/s, More Is BetterApache Cassandra 5.0Test: WritespxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4160K120K180K240K300K1739462713332382272358932368261749601740251736331391106651966243

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.3Blend File: Fishy Cat - Compute: CPU-OnlypxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b416012018024030097.0971.3573.8673.9673.4896.67105.24105.45136.51282.19280.26

Apache CouchDB

OpenBenchmarking.orgSeconds, Fewer Is BetterApache CouchDB 3.4.1Bulk Size: 300 - Inserts: 1000 - Rounds: 30pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41306090120150119.35106.13116.62116.24116.22117.57116.42118.16118.32154.04153.031. (CXX) g++ options: -flto -lstdc++ -shared -lei

PyPerformance

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.11Benchmark: async_tree_iopxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b412004006008001000656755651653633666742738726858855

OpenVINO GenAI

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO GenAI 2024.5Model: Gemma-7b-int4-ov - Device: CPU - Time Per Output TokenpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b30609012015097.61101.7298.6498.9198.3197.7998.0398.33105.09124.24

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO GenAI 2024.5Model: Gemma-7b-int4-ov - Device: CPU - Time To First TokenpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b60120180240300122.30106.62104.38105.36104.87121.48124.52124.38154.83288.64

OpenBenchmarking.orgtokens/s, More Is BetterOpenVINO GenAI 2024.5Model: Gemma-7b-int4-ov - Device: CPUpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b369121510.249.8310.1410.1110.1710.2310.2010.179.528.05

Llamafile

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.8.16Model: Llama-3.2-3B-Instruct.Q6_K - Test: Prompt Processing 2048pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b7K14K21K28K35K32768327683276832768327683276832768327683276832768

Rustls

OpenBenchmarking.orghandshakes/s, More Is BetterRustls 0.23.17Benchmark: handshake - Suite: TLS_ECDHE_ECDSA_WITH_AES_256_GCM_SHA384pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4190K180K270K360K450K304060.28423535.68436101.34435194.20435926.40306153.20301217.61301017.23225379.66105324.92106833.781. (CC) gcc options: -m64 -lgcc_s -lutil -lrt -lpthread -lm -ldl -lc -pie -nodefaultlibs

Llamafile

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.8.16Model: wizardcoder-python-34b-v1.0.Q6_K - Test: Prompt Processing 512pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b70014002100280035003072307230723072307230723072307230723072

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.3Encoder Mode: Preset 5 - Input: Bosphorus 4KpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4181624324028.8234.5434.4134.1634.6629.0927.5627.3619.8510.8110.871. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

OSPRay

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 3.2Benchmark: particle_volume/ao/real_timepxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4136912156.522069.009178.571468.515738.548786.527766.292116.288074.639652.266122.30223

CP2K Molecular Dynamics

OpenBenchmarking.orgSeconds, Fewer Is BetterCP2K Molecular Dynamics 2024.3Input: Fayalite-FISTpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41408012016020094.9094.0384.2184.8084.0792.2197.45100.49108.44179.12177.141. (F9X) gfortran options: -fopenmp -march=native -mtune=native -O3 -funroll-loops -fbacktrace -ffree-form -fimplicit-none -std=f2008 -lcp2kstart -lcp2kmc -lcp2kswarm -lcp2kmotion -lcp2kthermostat -lcp2kemd -lcp2ktmc -lcp2kmain -lcp2kdbt -lcp2ktas -lcp2kgrid -lcp2kgriddgemm -lcp2kgridcpu -lcp2kgridref -lcp2kgridcommon -ldbcsrarnoldi -ldbcsrx -lcp2kdbx -lcp2kdbm -lcp2kshg_int -lcp2keri_mme -lcp2kminimax -lcp2khfxbase -lcp2ksubsys -lcp2kxc -lcp2kao -lcp2kpw_env -lcp2kinput -lcp2kpw -lcp2kgpu -lcp2kfft -lcp2kfpga -lcp2kfm -lcp2kcommon -lcp2koffload -lcp2kmpiwrap -lcp2kbase -ldbcsr -lsirius -lspla -lspfft -lsymspg -lvdwxc -l:libhdf5_fortran.a -l:libhdf5.a -lz -lgsl -lelpa_openmp -lcosma -lcosta -lscalapack -lxsmmf -lxsmm -ldl -lpthread -llibgrpp -lxcf03 -lxc -lint2 -lfftw3_mpi -lfftw3 -lfftw3_omp -lmpi_cxx -lmpi -l:libopenblas.a -lvori -lstdc++ -lmpi_usempif08 -lmpi_mpifh -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm

Whisper.cpp

OpenBenchmarking.orgSeconds, Fewer Is BetterWhisper.cpp 1.6.2Model: ggml-base.en - Input: 2016 State of the UnionpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41408012016020093.4587.4984.0384.6884.1192.7197.5597.94103.72186.23187.291. (CXX) g++ options: -O3 -std=c++11 -fPIC -pthread -msse3 -mssse3 -mavx -mf16c -mfma -mavx2 -mavx512f -mavx512cd -mavx512vl -mavx512dq -mavx512bw -mavx512vbmi -mavx512vnni

PyPerformance

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.11Benchmark: asyncio_tcp_sslpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41160320480640800590645586582581590628626610737723

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.11Benchmark: xml_etreepxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4181624324036.535.836.236.136.136.835.635.334.736.636.7

Numpy Benchmark

This is a test to obtain the general Numpy performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterNumpy BenchmarkpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b412004006008001000831.42775.75754.41806.83833.64745.59783.43781.66813.31700.76705.99

simdjson

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 3.10Throughput Test: KostyapxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b412468105.455.976.046.075.546.116.025.786.065.685.701. (CXX) g++ options: -O3 -lrt

Llama.cpp

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b4154Backend: CPU BLAS - Model: Llama-3.1-Tulu-3-8B-Q8_0 - Test: Prompt Processing 1024pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b2040608010066.3570.8591.3190.0190.9666.5763.5063.7159.2731.911. (CXX) g++ options: -std=c++11 -fPIC -O3 -pthread -fopenmp -march=native -mtune=native -lopenblas

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b4154Backend: CPU BLAS - Model: Mistral-7B-Instruct-v0.3-Q8_0 - Test: Prompt Processing 1024pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b2040608010066.5269.2690.4891.1291.1766.8562.8464.1559.5432.591. (CXX) g++ options: -std=c++11 -fPIC -O3 -pthread -fopenmp -march=native -mtune=native -lopenblas

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.3Blend File: BMW27 - Compute: CPU-OnlypxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b415010015020025073.1653.5555.7455.2655.1874.0879.2479.19103.97214.89215.53

Llama.cpp

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b4154Backend: CPU BLAS - Model: Llama-3.1-Tulu-3-8B-Q8_0 - Test: Text Generation 128pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b2468107.126.887.007.007.007.117.107.096.646.551. (CXX) g++ options: -std=c++11 -fPIC -O3 -pthread -fopenmp -march=native -mtune=native -lopenblas

Llamafile

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.8.16Model: Llama-3.2-3B-Instruct.Q6_K - Test: Text Generation 128pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b51015202520.5120.1320.2620.2320.2920.3920.5020.4819.6020.08

NAMD

OpenBenchmarking.orgns/day, More Is BetterNAMD 3.0Input: STMV with 1,066,628 AtomspxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b410.17150.3430.51450.6860.85750.654480.756560.756450.759870.762270.651190.610930.614520.497710.273200.27352

Llama.cpp

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b4154Backend: CPU BLAS - Model: Mistral-7B-Instruct-v0.3-Q8_0 - Test: Text Generation 128pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b2468107.447.247.377.327.377.417.447.427.026.901. (CXX) g++ options: -std=c++11 -fPIC -O3 -pthread -fopenmp -march=native -mtune=native -lopenblas

OpenVINO GenAI

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO GenAI 2024.5Model: Falcon-7b-instruct-int4-ov - Device: CPU - Time Per Output TokenpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b2040608010074.5477.3475.5175.0674.8674.6575.0674.9479.4187.90

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO GenAI 2024.5Model: Falcon-7b-instruct-int4-ov - Device: CPU - Time To First TokenpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b5010015020025093.0086.0686.2086.0487.3493.0194.7793.00115.73217.85

OpenBenchmarking.orgtokens/s, More Is BetterOpenVINO GenAI 2024.5Model: Falcon-7b-instruct-int4-ov - Device: CPUpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b369121513.4112.9313.2413.3213.3613.4013.3213.3412.5911.38

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.3Encoder Mode: Preset 13 - Input: Beauty 4K 10-bitpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4151015202517.35518.58817.86317.84517.83617.40617.94418.0029.1887.9848.0071. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Stockfish

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 17Chess BenchmarkpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4113M26M39M52M65M42973396547527965867663657247763557627024526754642042951431588533151493714943737126302641. (CXX) g++ options: -lgcov -m64 -lpthread -fno-exceptions -std=c++17 -fno-peel-loops -fno-tracer -pedantic -O3 -funroll-loops -msse -msse3 -mpopcnt -mavx2 -mbmi -mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl -msse4.1 -mssse3 -msse2 -mbmi2 -flto -flto-partition=one -flto=jobserver

PyPerformance

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.11Benchmark: python_startuppxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b412468106.095.776.016.016.036.085.815.806.346.976.93

simdjson

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 3.10Throughput Test: LargeRandompxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b410.4140.8281.2421.6562.071.841.831.831.821.831.841.731.811.811.711.701. (CXX) g++ options: -O3 -lrt

ACES DGEMM

OpenBenchmarking.orgGFLOP/s, More Is BetterACES DGEMM 1.0Sustained Floating-Point RatepxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b412004006008001000842.011141.191093.101091.011092.09842.73812.15818.68629.17263.20259.151. (CC) gcc options: -ffast-math -mavx2 -O3 -fopenmp -lopenblas

Timed Eigen Compilation

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Eigen Compilation 3.4.0Time To CompilepxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4130609012015067.0858.6658.6958.8257.1867.3669.5869.3884.48146.66145.21

Apache CouchDB

OpenBenchmarking.orgSeconds, Fewer Is BetterApache CouchDB 3.4.1Bulk Size: 100 - Inserts: 1000 - Rounds: 30pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b412040608010076.3969.9376.6775.8976.6675.9076.5876.7073.2496.5596.741. (CXX) g++ options: -flto -lstdc++ -shared -lei

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.6Harness: Recurrent Neural Network Training - Engine: CPUpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41100020003000400050001895.681372.031320.371320.991320.231898.361913.861911.442422.834752.854726.08MIN: 1892.59MIN: 1342.06MIN: 1301.96MIN: 1308.56MIN: 1302.81MIN: 1894.26MIN: 1899.63MIN: 1895MIN: 2416.25MIN: 4746.78MIN: 4716.281. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -fcf-protection=full -pie -ldl

Renaissance

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: ALS Movie LenspxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b412K4K6K8K10K9275.79805.79969.89891.49972.09378.88980.89243.05993.47147.47465.1MIN: 8821.09 / MAX: 9495.91MIN: 9253.4 / MAX: 10057.61MIN: 9680.91 / MAX: 9983.16MIN: 9364.27 / MAX: 10037.94MIN: 9479.38 / MAX: 10040.33MIN: 8718.36 / MAX: 9413.7MIN: 8480.95 / MAX: 9113.57MIN: 8920.42 / MAX: 9406.46MIN: 5836.57 / MAX: 6535.35MIN: 6653.82 / MAX: 7863.42MIN: 6716.72 / MAX: 7815.35

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.6Harness: Recurrent Neural Network Inference - Engine: CPUpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b415001000150020002500966.01700.86670.23672.51670.70965.021012.191009.581270.762441.272415.25MIN: 963.43MIN: 679.89MIN: 663.39MIN: 665.03MIN: 662.66MIN: 963.27MIN: 999.17MIN: 994.85MIN: 1266.25MIN: 2435.83MIN: 2406.211. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -fcf-protection=full -pie -ldl

simdjson

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 3.10Throughput Test: DistinctUserIDpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4136912158.9710.4611.1411.059.1510.7610.2810.5410.8510.169.921. (CXX) g++ options: -O3 -lrt

Stockfish

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfishChess BenchmarkpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4110M20M30M40M50M33871595465070384572479444446458487970793370229829940074303159722401702411607502120246821. Stockfish 16 by the Stockfish developers (see AUTHORS file)

simdjson

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 3.10Throughput Test: TopTweetpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41369121510.5110.469.1911.259.4010.8210.6910.6610.869.9710.021. (CXX) g++ options: -O3 -lrt

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 3.10Throughput Test: PartialTweetspxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4136912158.359.769.989.8510.3210.109.639.8010.049.309.321. (CXX) g++ options: -O3 -lrt

Llamafile

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.8.16Model: TinyLlama-1.1B-Chat-v1.0.BF16 - Test: Text Generation 128pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b71421283527.8026.2827.1927.2027.2127.5927.3527.3625.5225.09

Renaissance

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: In-Memory Database ShootoutpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4180016002400320040003175.63256.13442.93232.83408.43241.53275.53288.53244.83681.92951.1MIN: 2896.06 / MAX: 3367.44MIN: 3019.89 / MAX: 3599.5MIN: 3258.62 / MAX: 3709.4MIN: 3057.86 / MAX: 3585.74MIN: 3187.55 / MAX: 3638.98MIN: 3037.03 / MAX: 3491.91MIN: 3012.8 / MAX: 3533.02MIN: 2991.78 / MAX: 3586.63MIN: 2350.89MIN: 2559.5MIN: 2787.28 / MAX: 3070.94

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: Akka Unbalanced Cobwebbed TreepxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41100020003000400050004002.34403.84344.84384.04359.34038.44098.54084.63720.14457.14474.8MIN: 4002.27 / MAX: 4983.72MAX: 5719.11MAX: 5622.35MIN: 4383.98 / MAX: 5691.67MIN: 4359.25 / MAX: 5618.71MIN: 4038.36 / MAX: 5089.28MIN: 4098.48 / MAX: 5163.21MAX: 5256.95MIN: 3720.09 / MAX: 4686.78MIN: 4457.08 / MAX: 5796.57MIN: 4474.77 / MAX: 5751.45

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: Apache Spark PageRankpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4150010001500200025002229.72412.22227.92182.52179.02138.12373.02366.12506.82292.22267.1MIN: 1612.96 / MAX: 2229.74MIN: 1691.04MIN: 1592.13 / MAX: 2227.91MIN: 1564.17MIN: 1591.55 / MAX: 2179.02MIN: 1499.64MIN: 1684.52MIN: 1667.92 / MAX: 2366.13MIN: 1771.47MIN: 2106.39 / MAX: 2374.31MIN: 2117.39 / MAX: 2335.98

Llamafile

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.8.16Model: mistral-7b-instruct-v0.2.Q5_K_M - Test: Prompt Processing 512pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b2K4K6K8K10K8192819281928192819281928192819281928192

CP2K Molecular Dynamics

OpenBenchmarking.orgSeconds, Fewer Is BetterCP2K Molecular Dynamics 2024.3Input: H20-64pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4130609012015052.7258.1942.5143.1742.4753.0158.8658.4068.17130.81123.661. (F9X) gfortran options: -fopenmp -march=native -mtune=native -O3 -funroll-loops -fbacktrace -ffree-form -fimplicit-none -std=f2008 -lcp2kstart -lcp2kmc -lcp2kswarm -lcp2kmotion -lcp2kthermostat -lcp2kemd -lcp2ktmc -lcp2kmain -lcp2kdbt -lcp2ktas -lcp2kgrid -lcp2kgriddgemm -lcp2kgridcpu -lcp2kgridref -lcp2kgridcommon -ldbcsrarnoldi -ldbcsrx -lcp2kdbx -lcp2kdbm -lcp2kshg_int -lcp2keri_mme -lcp2kminimax -lcp2khfxbase -lcp2ksubsys -lcp2kxc -lcp2kao -lcp2kpw_env -lcp2kinput -lcp2kpw -lcp2kgpu -lcp2kfft -lcp2kfpga -lcp2kfm -lcp2kcommon -lcp2koffload -lcp2kmpiwrap -lcp2kbase -ldbcsr -lsirius -lspla -lspfft -lsymspg -lvdwxc -l:libhdf5_fortran.a -l:libhdf5.a -lz -lgsl -lelpa_openmp -lcosma -lcosta -lscalapack -lxsmmf -lxsmm -ldl -lpthread -llibgrpp -lxcf03 -lxc -lint2 -lfftw3_mpi -lfftw3 -lfftw3_omp -lmpi_cxx -lmpi -l:libopenblas.a -lvori -lstdc++ -lmpi_usempif08 -lmpi_mpifh -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm

Renaissance

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: Savina Reactors.IOpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4180016002400320040003676.03506.43641.93582.43589.43655.83688.63523.13349.53818.53745.8MAX: 4536.84MIN: 3506.38 / MAX: 4329.37MAX: 4585.77MIN: 3582.35 / MAX: 4689.63MAX: 4472.27MIN: 3655.76 / MAX: 4484.97MAX: 4840.82MIN: 3523.09 / MAX: 4370.41MIN: 3349.49 / MAX: 4130.17MAX: 4702.95MIN: 3745.79 / MAX: 4547.1

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: Gaussian Mixture ModelpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4180016002400320040003815.23399.53773.13806.53829.93860.63703.23647.23397.03130.43087.0MIN: 2749.56 / MAX: 3815.24MIN: 2471.52MIN: 2755.26 / MAX: 3773.12MIN: 2770.53 / MAX: 3806.52MIN: 2792.29MIN: 2758.89 / MAX: 3860.61MIN: 2648.51 / MAX: 3703.23MIN: 2576.86 / MAX: 3647.22MIN: 2497.54 / MAX: 3397.03MIN: 2997.45 / MAX: 3481.57MIN: 2935.02 / MAX: 3439.02

PyPerformance

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.11Benchmark: gc_collectpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41150300450600750706677686682689699671671649704702

Renaissance

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: Apache Spark BayespxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41130260390520650474.9490.0515.6477.0500.0513.2513.3515.3483.4600.1566.2MIN: 454.77 / MAX: 514.32MIN: 459.29 / MAX: 580.9MIN: 459.35 / MAX: 536.06MIN: 460.78 / MAX: 515.59MIN: 457.03 / MAX: 581.14MIN: 453.66 / MAX: 554.7MIN: 455.04 / MAX: 535.53MIN: 456.38 / MAX: 531.31MIN: 456.88 / MAX: 535.94MIN: 506.71 / MAX: 741.31MIN: 493.35 / MAX: 1029.04

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: Finagle HTTP RequestspxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4160012001800240030002483.12319.42571.22627.32570.52492.22708.12650.91705.51669.91664.4MIN: 1933.43MIN: 1832.84MIN: 1999.64MIN: 2034.59 / MAX: 2627.31MIN: 1960.33 / MAX: 2570.51MIN: 1947.63MIN: 2074.84 / MAX: 2708.11MIN: 2066.79MIN: 1676.49 / MAX: 1730.57MIN: 1612.75 / MAX: 1713.57MIN: 1625.17 / MAX: 1776.68

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: Random ForestpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41110220330440550453.2414.4402.2420.7407.7422.0412.9415.3381.6485.4439.9MIN: 352.31 / MAX: 513.31MIN: 322.79 / MAX: 466.1MIN: 343.06 / MAX: 486.63MIN: 339.2 / MAX: 478.26MIN: 337.66 / MAX: 471.92MIN: 357.91 / MAX: 497.55MIN: 342.95 / MAX: 473.15MIN: 342.21 / MAX: 486.26MIN: 332.96 / MAX: 445.77MIN: 403.24 / MAX: 525.59MIN: 388.86 / MAX: 516.9

OSPRay

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 3.2Benchmark: gravity_spheres_volume/dim_512/scivis/real_timepxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b412468105.614707.587897.322847.323537.315945.548885.361895.357563.831521.894691.90804

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 3.2Benchmark: gravity_spheres_volume/dim_512/ao/real_timepxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b412468105.710847.639447.445757.387607.463995.631225.470575.470433.917491.904811.84407

ONNX Runtime

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.19Model: ResNet101_DUC_HDC-12 - Device: CPU - Executor: StandardpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b415001000150020002500854.33648.52647.81650.96645.21850.14864.81872.741091.322152.922151.821. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.19Model: ResNet101_DUC_HDC-12 - Device: CPU - Executor: StandardpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b410.34870.69741.04611.39481.74351.1705001.5419601.5436501.5361901.5498801.1762701.1563201.1458100.9163170.4644850.4647231. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

Renaissance

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: Scala DottypxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41110220330440550436.2477.0411.9441.9455.6428.6442.9424.6464.5515.8506.5MIN: 380.62 / MAX: 721.56MIN: 371.54 / MAX: 736.5MIN: 362.36 / MAX: 713.71MIN: 387.7 / MAX: 646.01MIN: 394.05 / MAX: 695.67MIN: 378.22 / MAX: 628.77MIN: 376.04 / MAX: 787.17MIN: 376.26 / MAX: 659.55MIN: 358.07 / MAX: 790.35MIN: 426.08 / MAX: 1026.55MIN: 420.76 / MAX: 1038.72

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: Genetic Algorithm Using Jenetics + FuturespxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b412004006008001000920.7732.8959.31050.3891.2904.0884.0924.5749.9801.4806.4MIN: 888.75 / MAX: 934.44MIN: 713.67 / MAX: 813.49MIN: 844.95MIN: 1016.46 / MAX: 1068.02MIN: 861.29 / MAX: 903.79MIN: 886.83 / MAX: 919.31MIN: 863.46 / MAX: 897.46MIN: 821.03MIN: 737.7 / MAX: 777.92MIN: 786.46 / MAX: 836.72MIN: 786.25 / MAX: 832.34

ONNX Runtime

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.19Model: fcn-resnet101-11 - Device: CPU - Executor: StandardpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b412004006008001000357.60310.88294.93295.28293.96355.75386.06386.56487.30927.39926.761. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.19Model: fcn-resnet101-11 - Device: CPU - Executor: StandardpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b410.76541.53082.29623.06163.8272.796383.216703.390643.386583.401782.810932.590252.586882.052111.078291.079031. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.19Model: GPT-2 - Device: CPU - Executor: StandardpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b412468106.330347.427766.030246.020696.016316.258157.365447.333317.960698.107678.048351. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.19Model: GPT-2 - Device: CPU - Executor: StandardpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b414080120160200157.89134.60165.77166.05166.17159.71135.72136.30125.52123.27124.171. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OSPRay

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 3.2Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_timepxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b412468106.407408.820938.477278.437528.496466.411986.176246.173994.523182.223182.23851

ONNX Runtime

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.19Model: ZFNet-512 - Device: CPU - Executor: StandardpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41481216209.016879.769858.825818.874448.835249.013229.838939.8589610.2835013.6955013.532301. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.19Model: ZFNet-512 - Device: CPU - Executor: StandardpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41306090120150110.89102.33113.29112.67113.17110.94101.62101.4197.2373.0073.881. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.19Model: bertsquad-12 - Device: CPU - Executor: StandardpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b412040608010068.6164.1466.9266.6267.1868.9168.0968.0850.4794.1295.071. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.19Model: bertsquad-12 - Device: CPU - Executor: StandardpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4151015202514.5715.5914.9415.0114.8814.5114.6914.6919.8110.6210.521. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.19Model: T5 Encoder - Device: CPU - Executor: StandardpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b412468104.851426.391124.913814.902854.903134.802876.004705.950357.060757.138577.041221. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.19Model: T5 Encoder - Device: CPU - Executor: StandardpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4150100150200250206.09156.45203.49203.95203.93208.17166.51168.04141.57140.03141.961. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.19Model: yolov4 - Device: CPU - Executor: StandardpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4130609012015093.3490.4588.3688.0688.4193.1695.3695.2575.54139.72139.661. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.19Model: yolov4 - Device: CPU - Executor: StandardpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41369121510.7127011.0552011.3166011.3551011.3105010.7338010.4865010.4982013.237207.157127.160151. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.19Model: ArcFace ResNet-100 - Device: CPU - Executor: StandardpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41102030405026.9523.5522.7022.8122.9926.7527.4727.6825.5841.8241.941. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.19Model: ArcFace ResNet-100 - Device: CPU - Executor: StandardpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41102030405037.1042.4544.0543.8443.4937.3836.4036.1239.0823.9123.841. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.19Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: StandardpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4171421283523.0621.2420.8820.7220.9824.9421.6921.5722.3225.9527.691. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.19Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: StandardpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41112233445543.3647.0747.8948.2647.6740.0946.0946.3544.8038.5336.111. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.19Model: CaffeNet 12-int8 - Device: CPU - Executor: StandardpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b410.45430.90861.36291.81722.27151.0660001.5708400.9997010.9941930.9866431.0618801.3351601.3797001.3112102.0191302.0160901. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.19Model: CaffeNet 12-int8 - Device: CPU - Executor: StandardpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b412004006008001000937.78636.32999.971005.501013.19941.40748.66724.53762.25495.08495.841. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.19Model: ResNet50 v1-12-int8 - Device: CPU - Executor: StandardpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b411.11292.22583.33874.45165.56452.806952.558982.612212.609742.625732.805442.767142.774222.654824.946364.815301. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.19Model: ResNet50 v1-12-int8 - Device: CPU - Executor: StandardpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4180160240320400356.19390.60382.77383.13380.80356.41361.30360.35376.53202.12207.611. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.19Model: super-resolution-10 - Device: CPU - Executor: StandardpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41481216207.994867.086018.182658.185198.181757.988737.433727.410376.3656413.5417013.645701. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.19Model: super-resolution-10 - Device: CPU - Executor: StandardpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41306090120150125.08141.12122.21122.17122.22125.17134.52134.94157.0873.8473.271. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

Llama.cpp

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b4154Backend: CPU BLAS - Model: granite-3.0-3b-a800m-instruct-Q8_0 - Test: Prompt Processing 2048pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b60120180240300208.99279.04289.04289.41273.67222.75212.30213.02171.68105.101. (CXX) g++ options: -std=c++11 -fPIC -O3 -pthread -fopenmp -march=native -mtune=native -lopenblas

PyPerformance

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.11Benchmark: asyncio_websocketspxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4170140210280350322315317315314321316315308331330

Llamafile

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.8.16Model: Llama-3.2-3B-Instruct.Q6_K - Test: Prompt Processing 1024pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4K8K12K16K20K16384163841638416384163841638416384163841638416384

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.8.16Model: wizardcoder-python-34b-v1.0.Q6_K - Test: Prompt Processing 256pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b300600900120015001536153615361536153615361536153615361536

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.8.16Model: TinyLlama-1.1B-Chat-v1.0.BF16 - Test: Prompt Processing 2048pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b7K14K21K28K35K32768327683276832768327683276832768327683276832768

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.3Encoder Mode: Preset 5 - Input: Bosphorus 1080ppxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b412040608010088.27101.9799.4598.7998.9188.4286.3986.9665.3237.0737.171. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

LiteRT

OpenBenchmarking.orgMicroseconds, Fewer Is BetterLiteRT 2024-10-15Model: Inception V4pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4111K22K33K44K55K22752.421477.819859.420102.819686.022083.322928.323141.424502.149101.949048.8

OpenBenchmarking.orgMicroseconds, Fewer Is BetterLiteRT 2024-10-15Model: Inception ResNet V2pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b419K18K27K36K45K19490.719530.218412.318224.017824.619477.820503.520123.521153.541757.941844.0

OpenBenchmarking.orgMicroseconds, Fewer Is BetterLiteRT 2024-10-15Model: NASNet MobilepxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b414K8K12K16K20K7931.6416936.009231.979311.779245.768057.567640.817759.693808.805034.385153.49

OpenBenchmarking.orgMicroseconds, Fewer Is BetterLiteRT 2024-10-15Model: DeepLab V3pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4180016002400320040002359.993579.672200.932185.782152.482343.382289.582272.281569.752994.592950.80

OpenBenchmarking.orgMicroseconds, Fewer Is BetterLiteRT 2024-10-15Model: Mobilenet FloatpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4150010001500200025001244.511211.481175.501165.501178.941244.701248.601248.121044.532190.792193.88

OpenBenchmarking.orgMicroseconds, Fewer Is BetterLiteRT 2024-10-15Model: SqueezeNetpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4170014002100280035001821.351794.111765.391769.561763.221809.181822.701820.341654.653179.553173.55

OpenBenchmarking.orgMicroseconds, Fewer Is BetterLiteRT 2024-10-15Model: Quantized COCO SSD MobileNet v1pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4160012001800240030001417.352129.521339.691337.781343.221420.151415.111431.191494.512741.682736.98

OpenBenchmarking.orgMicroseconds, Fewer Is BetterLiteRT 2024-10-15Model: Mobilenet QuantpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41400800120016002000849.21823.17742.03734.35730.17848.94857.49859.76935.991785.401781.60

Rustls

OpenBenchmarking.orghandshakes/s, More Is BetterRustls 0.23.17Benchmark: handshake-ticket - Suite: TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41600K1200K1800K2400K3000K2292879.442620332.002558621.382528614.522559245.202282729.642304145.792283083.542353677.961571309.621656876.391. (CC) gcc options: -m64 -lgcc_s -lutil -lrt -lpthread -lm -ldl -lc -pie -nodefaultlibs

Llama.cpp

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b4154Backend: CPU BLAS - Model: Llama-3.1-Tulu-3-8B-Q8_0 - Test: Prompt Processing 512pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b2040608010067.9570.7693.9192.8991.7169.1164.4364.9660.8732.971. (CXX) g++ options: -std=c++11 -fPIC -O3 -pthread -fopenmp -march=native -mtune=native -lopenblas

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b4154Backend: CPU BLAS - Model: Mistral-7B-Instruct-v0.3-Q8_0 - Test: Prompt Processing 512pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b2040608010068.8168.4093.4993.2494.0868.2064.6865.2461.0932.941. (CXX) g++ options: -std=c++11 -fPIC -O3 -pthread -fopenmp -march=native -mtune=native -lopenblas

Rustls

OpenBenchmarking.orghandshakes/s, More Is BetterRustls 0.23.17Benchmark: handshake-resume - Suite: TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41800K1600K2400K3200K4000K3038723.483563852.573431395.733397871.033379343.723035330.213082657.503061522.262874011.562103587.502105346.651. (CC) gcc options: -m64 -lgcc_s -lutil -lrt -lpthread -lm -ldl -lc -pie -nodefaultlibs

ASTC Encoder

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 5.0Preset: ThoroughpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4151015202514.146420.302518.749918.707518.766114.170013.513913.523010.33925.04485.06551. (CXX) g++ options: -O3 -flto -pthread

Llamafile

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.8.16Model: wizardcoder-python-34b-v1.0.Q6_K - Test: Text Generation 16pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b0.41850.8371.25551.6742.09251.841.781.831.821.861.831.831.831.761.73

FinanceBench

FinanceBench is a collection of financial program benchmarks with support for benchmarking on the GPU via OpenCL and CPU benchmarking with OpenMP. The FinanceBench test cases are focused on Black-Sholes-Merton Process with Analytic European Option engine, QMC (Sobol) Monte-Carlo method (Equity Option Example), Bonds Fixed-rate bond with flat forward curve, and Repo Securities repurchase agreement. FinanceBench was originally written by the Cavazos Lab at University of Delaware. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Bonds OpenMPpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4110K20K30K40K50K34896.8433061.2233913.4634053.0633906.6934600.7733538.2733551.5232715.8146851.4247198.941. (CXX) g++ options: -O3 -march=native -fopenmp

Etcpak

OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 2.0Benchmark: Multi-Threaded - Configuration: ETC2pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41120240360480600409.88577.82556.41556.04556.87410.73388.13388.40304.24154.25155.441. (CXX) g++ options: -flto -pthread

Whisperfile

OpenBenchmarking.orgSeconds, Fewer Is BetterWhisperfile 20Aug24Model Size: TinypxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b153045607538.7241.7135.4335.9135.1137.1338.5840.8541.3066.11

Rustls

OpenBenchmarking.orghandshakes/s, More Is BetterRustls 0.23.17Benchmark: handshake - Suite: TLS13_CHACHA20_POLY1305_SHA256pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4116K32K48K64K80K57688.0876454.4572508.1372793.2672140.7557716.6456176.9056047.3843571.8621444.0921499.731. (CC) gcc options: -m64 -lgcc_s -lutil -lrt -lpthread -lm -ldl -lc -pie -nodefaultlibs

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.3Encoder Mode: Preset 8 - Input: Bosphorus 4KpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b412040608010085.00102.01101.54101.33101.5285.2082.1682.5365.1935.6435.851. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Rustls

OpenBenchmarking.orghandshakes/s, More Is BetterRustls 0.23.17Benchmark: handshake - Suite: TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4120K40K60K80K100K59206.3480462.6074683.7474468.4974672.5759308.7558056.0157954.1945095.3522201.0322255.231. (CC) gcc options: -m64 -lgcc_s -lutil -lrt -lpthread -lm -ldl -lc -pie -nodefaultlibs

PyPerformance

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.11Benchmark: django_templatepxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4151015202521.220.721.020.820.921.020.620.620.121.321.2

OpenVINO GenAI

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO GenAI 2024.5Model: Phi-3-mini-128k-instruct-int4-ov - Device: CPU - Time Per Output TokenpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b132639526549.2851.8650.6850.8450.7449.3150.4650.4552.6460.07

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO GenAI 2024.5Model: Phi-3-mini-128k-instruct-int4-ov - Device: CPU - Time To First TokenpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b30609012015058.8655.9355.4756.0655.6758.9159.8659.8771.58130.47

OpenBenchmarking.orgtokens/s, More Is BetterOpenVINO GenAI 2024.5Model: Phi-3-mini-128k-instruct-int4-ov - Device: CPUpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b51015202520.2919.2819.7319.6719.7120.2819.8219.8219.0016.65

POV-Ray

OpenBenchmarking.orgSeconds, Fewer Is BetterPOV-RayTrace TimepxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41163248648025.3318.5420.3319.5519.5225.2626.6026.6134.9270.7770.061. POV-Ray 3.7.0.10.unofficial

NAMD

OpenBenchmarking.orgns/day, More Is BetterNAMD 3.0Input: ATPase with 327,506 AtomspxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b410.64231.28461.92692.56923.21152.353792.796322.854852.839132.837982.381242.207282.216451.740380.892410.88895

Llamafile

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.8.16Model: mistral-7b-instruct-v0.2.Q5_K_M - Test: Prompt Processing 256pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b90018002700360045004096409640964096409640964096409640964096

PyPerformance

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.11Benchmark: raytracepxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b414080120160200182175176176179182176177174180182

FinanceBench

FinanceBench is a collection of financial program benchmarks with support for benchmarking on the GPU via OpenCL and CPU benchmarking with OpenMP. The FinanceBench test cases are focused on Black-Sholes-Merton Process with Analytic European Option engine, QMC (Sobol) Monte-Carlo method (Equity Option Example), Bonds Fixed-rate bond with flat forward curve, and Repo Securities repurchase agreement. FinanceBench was originally written by the Cavazos Lab at University of Delaware. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Repo OpenMPpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b417K14K21K28K35K22318.7421418.4521848.0721948.9422060.1522320.3321612.0421588.9121121.6132477.3532271.671. (CXX) g++ options: -O3 -march=native -fopenmp

PyPerformance

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.11Benchmark: crypto_pyaespxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41102030405043.341.742.742.742.443.142.041.840.743.143.0

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.11Benchmark: chaospxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4191827364539.438.239.139.238.839.738.639.038.340.239.9

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.11Benchmark: gopxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b412040608010079.477.879.678.478.178.676.078.375.378.778.4

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.11Benchmark: floatpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41122436486050.850.751.051.150.751.349.749.748.252.952.8

x265

OpenBenchmarking.orgFrames Per Second, More Is Betterx265Video Input: Bosphorus 4KpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4181624324026.9432.5732.2232.5932.2327.1625.8526.0918.8510.8910.911. x265 [info]: HEVC encoder version 3.5+1-f0c1022b6

PyPerformance

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.11Benchmark: regex_compilepxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41163248648072.569.869.871.470.471.770.069.568.172.772.6

Llamafile

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.8.16Model: Llama-3.2-3B-Instruct.Q6_K - Test: Prompt Processing 512pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b2K4K6K8K10K8192819281928192819281928192819281928192

Llama.cpp

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b4154Backend: CPU BLAS - Model: granite-3.0-3b-a800m-instruct-Q8_0 - Test: Prompt Processing 1024pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b80160240320400244.77355.09324.04290.10291.02232.26227.71235.01185.97115.671. (CXX) g++ options: -std=c++11 -fPIC -O3 -pthread -fopenmp -march=native -mtune=native -lopenblas

Y-Cruncher

OpenBenchmarking.orgSeconds, Fewer Is BetterY-Cruncher 0.8.5Pi Digits To Calculate: 1BpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4191827364518.3718.4917.5717.3417.3618.3819.2819.3623.8939.2540.56

PyPerformance

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.11Benchmark: pathlibpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b414812162014.414.214.114.114.214.414.114.113.514.614.6

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.11Benchmark: pickle_pure_pythonpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b414080120160200168165165166163169164164162171171

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.6Harness: Deconvolution Batch shapes_1d - Engine: CPUpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b412468103.406282.976123.063413.072193.059183.402933.425473.432873.786297.281397.26738MIN: 3.03MIN: 2.42MIN: 2.58MIN: 2.59MIN: 2.56MIN: 3.03MIN: 2.9MIN: 2.96MIN: 3.55MIN: 6.94MIN: 6.921. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -fcf-protection=full -pie -ldl

Llamafile

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.8.16Model: mistral-7b-instruct-v0.2.Q5_K_M - Test: Text Generation 16pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b369121510.4510.2210.3710.3010.3410.4510.4510.489.929.99

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.8.16Model: TinyLlama-1.1B-Chat-v1.0.BF16 - Test: Prompt Processing 1024pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4K8K12K16K20K16384163841638416384163841638416384163841638416384

PyPerformance

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.11Benchmark: json_loadspxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41369121512.512.112.212.512.312.412.012.011.812.412.4

ASTC Encoder

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 5.0Preset: FastpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4190180270360450277.30396.65368.57368.55369.03278.24263.37263.55205.0199.40100.371. (CXX) g++ options: -O3 -flto -pthread

PyPerformance

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.11Benchmark: nbodypxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41142842567059.259.058.859.059.659.558.657.256.860.060.4

7-Zip Compression

OpenBenchmarking.orgMIPS, More Is Better7-Zip CompressionTest: Decompression RatingpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4140K80K120K160K200K1256051659161659341645811661801256981185871185698822943210437811. 7-Zip 23.01 (x64) : Copyright (c) 1999-2023 Igor Pavlov : 2023-06-20

OpenBenchmarking.orgMIPS, More Is Better7-Zip CompressionTest: Compression RatingpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4140K80K120K160K200K14221316385916975017036116937314126313380413503310187253425540791. 7-Zip 23.01 (x64) : Copyright (c) 1999-2023 Igor Pavlov : 2023-06-20

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.3Encoder Mode: Preset 13 - Input: Bosphorus 4KpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4150100150200250194.02212.52218.53217.91217.96198.11190.34192.17156.6085.6085.951. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.3Encoder Mode: Preset 8 - Input: Bosphorus 1080ppxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4170140210280350286.96339.02327.26320.18325.95287.05281.89282.01215.21122.92123.791. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.6Harness: IP Shapes 1D - Engine: CPUpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b410.8621.7242.5863.4484.311.939131.125731.128011.124651.130121.938061.889191.887311.906863.828853.83117MIN: 1.91MIN: 1.03MIN: 1.1MIN: 1.09MIN: 1.1MIN: 1.92MIN: 1.83MIN: 1.83MIN: 1.86MIN: 3.77MIN: 3.771. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -fcf-protection=full -pie -ldl

ASTC Encoder

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 5.0Preset: MediumpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41306090120150108.86156.22144.63144.48144.61109.03103.67103.8180.3739.0439.321. (CXX) g++ options: -O3 -flto -pthread

Llama.cpp

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b4154Backend: CPU BLAS - Model: granite-3.0-3b-a800m-instruct-Q8_0 - Test: Text Generation 128pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b122436486052.3747.7249.7649.9050.3352.3050.4350.5051.8651.501. (CXX) g++ options: -std=c++11 -fPIC -O3 -pthread -fopenmp -march=native -mtune=native -lopenblas

Llamafile

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.8.16Model: Llama-3.2-3B-Instruct.Q6_K - Test: Prompt Processing 256pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b90018002700360045004096409640964096409640964096409640964096

Llama.cpp

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b4154Backend: CPU BLAS - Model: granite-3.0-3b-a800m-instruct-Q8_0 - Test: Prompt Processing 512pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b70140210280350232.86327.30326.16316.80337.06243.14242.10235.94204.06122.801. (CXX) g++ options: -std=c++11 -fPIC -O3 -pthread -fopenmp -march=native -mtune=native -lopenblas

Llamafile

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.8.16Model: Llama-3.2-3B-Instruct.Q6_K - Test: Text Generation 16pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b51015202519.5019.0319.2419.2719.1619.4919.5319.4718.8319.17

Y-Cruncher

OpenBenchmarking.orgSeconds, Fewer Is BetterY-Cruncher 0.8.5Pi Digits To Calculate: 500MpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b415101520258.6238.7728.2798.1738.1198.6889.0679.04110.95418.29018.165

Primesieve

OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 12.6Length: 1e12pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b416121824309.1476.3476.9036.9266.9379.1169.7059.70112.27324.82124.7231. (CXX) g++ options: -O3

Llamafile

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.8.16Model: TinyLlama-1.1B-Chat-v1.0.BF16 - Test: Prompt Processing 512pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b2K4K6K8K10K8192819281928192819281928192819281928192

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.8.16Model: TinyLlama-1.1B-Chat-v1.0.BF16 - Test: Text Generation 16pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b61218243025.9424.5925.3025.3225.3825.8625.7825.7724.2723.80

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.6Harness: IP Shapes 3D - Engine: CPUpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b411.11572.23143.34714.46285.57852.729424.058002.777272.746992.778352.730723.745783.750034.450964.958674.94689MIN: 2.7MIN: 3.75MIN: 2.75MIN: 2.72MIN: 2.74MIN: 2.7MIN: 3.71MIN: 3.71MIN: 4.38MIN: 4.87MIN: 4.851. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -fcf-protection=full -pie -ldl

Renaissance

Test: Apache Spark ALS

a: The test quit with a non-zero exit status.

4484PX: The test quit with a non-zero exit status.

px: The test quit with a non-zero exit status.

4464p: The test quit with a non-zero exit status.

4464p epyc: The test quit with a non-zero exit status.

4364P: The test quit with a non-zero exit status.

4584PX: The test quit with a non-zero exit status.

EPYC 4584PX amd: The test quit with a non-zero exit status.

45: The test quit with a non-zero exit status.

41: The test quit with a non-zero exit status.

41 b: The test quit with a non-zero exit status.

x265

OpenBenchmarking.orgFrames Per Second, More Is Betterx265Video Input: Bosphorus 1080ppxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b41306090120150101.25114.45113.16113.37113.31101.37101.33101.2388.7156.0356.611. x265 [info]: HEVC encoder version 3.5+1-f0c1022b6

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.6Harness: Convolution Batch Shapes Auto - Engine: CPUpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b4136912154.133216.672874.048774.038084.035504.115516.371506.364757.873679.169569.18658MIN: 4.07MIN: 6.2MIN: 3.99MIN: 3.98MIN: 3.98MIN: 4.05MIN: 6.28MIN: 6.28MIN: 7.77MIN: 9.03MIN: 9.061. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -fcf-protection=full -pie -ldl

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.3Encoder Mode: Preset 13 - Input: Bosphorus 1080ppxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b412004006008001000769.82842.56845.58842.34835.45776.12748.15752.16652.97377.79380.631. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Llamafile

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.8.16Model: TinyLlama-1.1B-Chat-v1.0.BF16 - Test: Prompt Processing 256pxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b90018002700360045004096409640964096409640964096409640964096

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.6Harness: Deconvolution Batch shapes_3d - Engine: CPUpxaEPYC 4584PX amd4584PX454484PX4464p epyc4464p4364P41 b412468103.512432.412942.664292.672192.662963.508403.388233.389583.780087.553157.53655MIN: 3.47MIN: 2.34MIN: 2.62MIN: 2.62MIN: 2.62MIN: 3.46MIN: 3.28MIN: 3.28MIN: 3.64MIN: 7.52MIN: 7.531. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -fcf-protection=full -pie -ldl

OpenVINO GenAI

Model: TinyLlama-1.1B-Chat-v1.0 - Device: CPU

a: The test quit with a non-zero exit status. E: RuntimeError: Exception from src/inference/src/cpp/core.cpp:90:

4484PX: The test quit with a non-zero exit status. E: RuntimeError: Exception from src/inference/src/cpp/core.cpp:90:

px: The test quit with a non-zero exit status. E: RuntimeError: Exception from src/inference/src/cpp/core.cpp:90:

4464p: The test quit with a non-zero exit status. E: RuntimeError: Exception from src/inference/src/cpp/core.cpp:90:

4464p epyc: The test quit with a non-zero exit status. E: RuntimeError: Exception from src/inference/src/cpp/core.cpp:90:

4364P: The test quit with a non-zero exit status. E: RuntimeError: Exception from src/inference/src/cpp/core.cpp:90:

4584PX: The test quit with a non-zero exit status. E: RuntimeError: Exception from src/inference/src/cpp/core.cpp:90:

EPYC 4584PX amd: The test quit with a non-zero exit status. E: RuntimeError: Exception from src/inference/src/cpp/core.cpp:90:

45: The test quit with a non-zero exit status. E: RuntimeError: Exception from src/inference/src/cpp/core.cpp:90:

41 b: The test quit with a non-zero exit status. E: RuntimeError: Exception from src/inference/src/cpp/core.cpp:90:

OpenSSL

Algorithm: RSA4096

a: The test quit with a non-zero exit status.

4484PX: The test quit with a non-zero exit status.

px: The test quit with a non-zero exit status.

4464p: The test quit with a non-zero exit status.

4464p epyc: The test quit with a non-zero exit status.

4364P: The test quit with a non-zero exit status.

4584PX: The test quit with a non-zero exit status.

EPYC 4584PX amd: The test quit with a non-zero exit status.

45: The test quit with a non-zero exit status.

41: The test quit with a non-zero exit status.

41 b: The test quit with a non-zero exit status.

Algorithm: SHA512

a: The test quit with a non-zero exit status.

4484PX: The test quit with a non-zero exit status.

px: The test quit with a non-zero exit status.

4464p: The test quit with a non-zero exit status.

4464p epyc: The test quit with a non-zero exit status.

4364P: The test quit with a non-zero exit status.

4584PX: The test quit with a non-zero exit status.

EPYC 4584PX amd: The test quit with a non-zero exit status.

45: The test quit with a non-zero exit status.

41: The test quit with a non-zero exit status.

41 b: The test quit with a non-zero exit status.

Algorithm: SHA256

a: The test quit with a non-zero exit status.

4484PX: The test quit with a non-zero exit status.

px: The test quit with a non-zero exit status.

4464p: The test quit with a non-zero exit status.

4464p epyc: The test quit with a non-zero exit status.

4364P: The test quit with a non-zero exit status.

4584PX: The test quit with a non-zero exit status.

EPYC 4584PX amd: The test quit with a non-zero exit status.

45: The test quit with a non-zero exit status.

41: The test quit with a non-zero exit status.

41 b: The test quit with a non-zero exit status.

213 Results Shown

QuantLib
SVT-AV1
RELION
Blender
Whisper.cpp
CP2K Molecular Dynamics
Apache CouchDB
QuantLib
Whisperfile
Llamafile
Apache CouchDB
SVT-AV1
BYTE Unix Benchmark
Whisper.cpp
Llamafile
Blender
Llamafile
SVT-AV1
Apache CouchDB
BYTE Unix Benchmark:
  Pipe
  Dhrystone 2
  System Call
Blender
Rustls:
  handshake-resume - TLS13_CHACHA20_POLY1305_SHA256
  handshake-ticket - TLS13_CHACHA20_POLY1305_SHA256
Llamafile
Llama.cpp:
  CPU BLAS - Llama-3.1-Tulu-3-8B-Q8_0 - Prompt Processing 2048
  CPU BLAS - Mistral-7B-Instruct-v0.3-Q8_0 - Prompt Processing 2048
XNNPACK:
  QS8MobileNetV2
  FP16MobileNetV3Small
  FP16MobileNetV3Large
  FP16MobileNetV2
  FP32MobileNetV3Small
  FP32MobileNetV3Large
  FP32MobileNetV2
  FP32MobileNetV1
  FP16MobileNetV1
Rustls
Whisperfile
Rustls
OpenSSL:
  ChaCha20
  ChaCha20-Poly1305
  AES-256-GCM
  AES-128-GCM
Llamafile
SVT-AV1
Apache CouchDB
Gcrypt Library
ASTC Encoder:
  Very Thorough
  Exhaustive
OSPRay:
  particle_volume/scivis/real_time
  particle_volume/pathtracer/real_time
SVT-AV1
GROMACS
Build2
Primesieve
Llamafile
Blender
Apache Cassandra
Blender
Apache CouchDB
PyPerformance
OpenVINO GenAI:
  Gemma-7b-int4-ov - CPU - Time Per Output Token
  Gemma-7b-int4-ov - CPU - Time To First Token
  Gemma-7b-int4-ov - CPU
Llamafile
Rustls
Llamafile
SVT-AV1
OSPRay
CP2K Molecular Dynamics
Whisper.cpp
PyPerformance:
  asyncio_tcp_ssl
  xml_etree
Numpy Benchmark
simdjson
Llama.cpp:
  CPU BLAS - Llama-3.1-Tulu-3-8B-Q8_0 - Prompt Processing 1024
  CPU BLAS - Mistral-7B-Instruct-v0.3-Q8_0 - Prompt Processing 1024
Blender
Llama.cpp
Llamafile
NAMD
Llama.cpp
OpenVINO GenAI:
  Falcon-7b-instruct-int4-ov - CPU - Time Per Output Token
  Falcon-7b-instruct-int4-ov - CPU - Time To First Token
  Falcon-7b-instruct-int4-ov - CPU
SVT-AV1
Stockfish
PyPerformance
simdjson
ACES DGEMM
Timed Eigen Compilation
Apache CouchDB
oneDNN
Renaissance
oneDNN
simdjson
Stockfish
simdjson:
  TopTweet
  PartialTweets
Llamafile
Renaissance:
  In-Memory Database Shootout
  Akka Unbalanced Cobwebbed Tree
  Apache Spark PageRank
Llamafile
CP2K Molecular Dynamics
Renaissance:
  Savina Reactors.IO
  Gaussian Mixture Model
PyPerformance
Renaissance:
  Apache Spark Bayes
  Finagle HTTP Requests
  Rand Forest
OSPRay:
  gravity_spheres_volume/dim_512/scivis/real_time
  gravity_spheres_volume/dim_512/ao/real_time
ONNX Runtime:
  ResNet101_DUC_HDC-12 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
Renaissance:
  Scala Dotty
  Genetic Algorithm Using Jenetics + Futures
ONNX Runtime:
  fcn-resnet101-11 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
  GPT-2 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
OSPRay
ONNX Runtime:
  ZFNet-512 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
  bertsquad-12 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
  T5 Encoder - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
  yolov4 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
  ArcFace ResNet-100 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
  Faster R-CNN R-50-FPN-int8 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
  CaffeNet 12-int8 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
  ResNet50 v1-12-int8 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
  super-resolution-10 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
Llama.cpp
PyPerformance
Llamafile:
  Llama-3.2-3B-Instruct.Q6_K - Prompt Processing 1024
  wizardcoder-python-34b-v1.0.Q6_K - Prompt Processing 256
  TinyLlama-1.1B-Chat-v1.0.BF16 - Prompt Processing 2048
SVT-AV1
LiteRT:
  Inception V4
  Inception ResNet V2
  NASNet Mobile
  DeepLab V3
  Mobilenet Float
  SqueezeNet
  Quantized COCO SSD MobileNet v1
  Mobilenet Quant
Rustls
Llama.cpp:
  CPU BLAS - Llama-3.1-Tulu-3-8B-Q8_0 - Prompt Processing 512
  CPU BLAS - Mistral-7B-Instruct-v0.3-Q8_0 - Prompt Processing 512
Rustls
ASTC Encoder
Llamafile
FinanceBench
Etcpak
Whisperfile
Rustls
SVT-AV1
Rustls
PyPerformance
OpenVINO GenAI:
  Phi-3-mini-128k-instruct-int4-ov - CPU - Time Per Output Token
  Phi-3-mini-128k-instruct-int4-ov - CPU - Time To First Token
  Phi-3-mini-128k-instruct-int4-ov - CPU
POV-Ray
NAMD
Llamafile
PyPerformance
FinanceBench
PyPerformance:
  crypto_pyaes
  chaos
  go
  float
x265
PyPerformance
Llamafile
Llama.cpp
Y-Cruncher
PyPerformance:
  pathlib
  pickle_pure_python
oneDNN
Llamafile:
  mistral-7b-instruct-v0.2.Q5_K_M - Text Generation 16
  TinyLlama-1.1B-Chat-v1.0.BF16 - Prompt Processing 1024
PyPerformance
ASTC Encoder
PyPerformance
7-Zip Compression:
  Decompression Rating
  Compression Rating
SVT-AV1:
  Preset 13 - Bosphorus 4K
  Preset 8 - Bosphorus 1080p
oneDNN
ASTC Encoder
Llama.cpp
Llamafile
Llama.cpp
Llamafile
Y-Cruncher
Primesieve
Llamafile:
  TinyLlama-1.1B-Chat-v1.0.BF16 - Prompt Processing 512
  TinyLlama-1.1B-Chat-v1.0.BF16 - Text Generation 16
oneDNN
x265
oneDNN
SVT-AV1
Llamafile
oneDNN