Linux 6.13 AMD EPYC Performance Benchmarks for a future article. AMD EPYC 9575F 64-Core testing with a Supermicro Super Server H13SSL-N v1.01 (3.0 BIOS) and ASPEED on Ubuntu 24.10 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2412073-NE-LINUX613A93&sro&grs .
Linux 6.13 AMD EPYC Performance Processor Motherboard Chipset Memory Disk Graphics Network OS Kernel Desktop Display Server Compiler File-System Screen Resolution v6.11 v6.12 v6.13 7 Dec AMD EPYC 9575F 64-Core @ 3.30GHz (64 Cores / 128 Threads) Supermicro Super Server H13SSL-N v1.01 (3.0 BIOS) AMD 1Ah 12 x 64GB DDR5-6000MT/s Micron MTC40F2046S1RC64BDY QSFF 3201GB Micron_7450_MTFDKCB3T2TFS ASPEED 2 x Broadcom NetXtreme BCM5720 PCIe Ubuntu 24.10 6.11.0-phx (x86_64) GNOME Shell 47.0 X Server GCC 14.2.0 ext4 1920x1200 6.12.0-phx (x86_64) 1024x768 6.13.0-rc1-phx (x86_64) OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise Compiler Details - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2,rust --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-14-zdkDXv/gcc-14-14.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-14-zdkDXv/gcc-14-14.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Processor Details - Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xb002116 Java Details - OpenJDK Runtime Environment (build 21.0.5+11-Ubuntu-1ubuntu124.10) Python Details - Python 3.12.7 Security Details - gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS; IBPB: conditional; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Linux 6.13 AMD EPYC Performance stress-ng: Context Switching mariadb: oltp_update_index - 128 dacapobench: Spring Boot dacapobench: Apache Lucene Search Engine stress-ng: NUMA dacapobench: Zxing 1D/2D Barcode Image Processing stress-ng: Semaphores pgbench: 100 - 1000 - Read Write - Average Latency pgbench: 100 - 1000 - Read Write mariadb: oltp_write_only - 128 memcached: 1:100 stress-ng: Futex stress-ng: MEMFD pgbench: 100 - 800 - Read Write - Average Latency stress-ng: Mutex pgbench: 100 - 800 - Read Write dacapobench: H2 Database Engine llamafile: mistral-7b-instruct-v0.2.Q5_K_M - Text Generation 128 couchdb: 300 - 1000 - 30 stress-ng: Mixed Scheduler dacapobench: Avrora AVR Simulation Framework svt-av1: Preset 13 - Bosphorus 4K pgbench: 100 - 1000 - Read Only - Average Latency llamafile: Llama-3.2-3B-Instruct.Q6_K - Text Generation 128 pgbench: 100 - 1000 - Read Only mariadb: oltp_read_write - 128 llamafile: TinyLlama-1.1B-Chat-v1.0.BF16 - Text Generation 128 mariadb: oltp_read_only - 128 laghos: Triple Point Problem memcached: 1:10 dacapobench: Apache Tomcat dacapobench: Batik SVG Toolkit dacapobench: Apache Lucene Search Index cassandra: Writes blender: Fishy Cat - CPU-Only clickhouse: 100M Rows Hits Dataset, First Run / Cold Cache dacapobench: Tradesoap blender: Classroom - CPU-Only dacapobench: BioJava Biological Data Framework memcached: 1:5 pgbench: 100 - 800 - Read Only - Average Latency dacapobench: FOP Print Formatter blender: BMW27 - CPU-Only dacapobench: Eclipse svt-av1: Preset 5 - Bosphorus 4K blender: Pabellon Barcelona - CPU-Only clickhouse: 100M Rows Hits Dataset, Second Run pgbench: 100 - 800 - Read Only openvino-genai: Falcon-7b-instruct-int4-ov - CPU dacapobench: GraphChi clickhouse: 100M Rows Hits Dataset, Third Run openssl: ChaCha20 openvino-genai: TinyLlama-1.1B-Chat-v1.0 - CPU openvino-genai: Gemma-7b-int4-ov - CPU llamafile: wizardcoder-python-34b-v1.0.Q6_K - Text Generation 128 svt-av1: Preset 8 - Bosphorus 4K blender: Barbershop - CPU-Only svt-av1: Preset 3 - Bosphorus 4K build-linux-kernel: allmodconfig dacapobench: Jython laghos: Sedov Blast Wave, ube_922_hex.mesh stress-ng: Socket Activity primesieve: 1e12 openssl: ChaCha20-Poly1305 dacapobench: Apache Kafka build-llvm: Unix Makefiles openssl: AES-128-GCM openssl: AES-256-GCM primesieve: 1e13 build-linux-kernel: defconfig dacapobench: Apache Xalan XSLT build-llvm: Ninja blender: Junkshop - CPU-Only openvino-genai: Phi-3-mini-128k-instruct-int4-ov - CPU dacapobench: jMonkeyEngine llamafile: wizardcoder-python-34b-v1.0.Q6_K - Prompt Processing 2048 llamafile: wizardcoder-python-34b-v1.0.Q6_K - Prompt Processing 1024 llamafile: wizardcoder-python-34b-v1.0.Q6_K - Prompt Processing 512 llamafile: wizardcoder-python-34b-v1.0.Q6_K - Prompt Processing 256 llamafile: mistral-7b-instruct-v0.2.Q5_K_M - Prompt Processing 2048 llamafile: mistral-7b-instruct-v0.2.Q5_K_M - Prompt Processing 1024 llamafile: mistral-7b-instruct-v0.2.Q5_K_M - Prompt Processing 512 llamafile: mistral-7b-instruct-v0.2.Q5_K_M - Prompt Processing 256 llamafile: TinyLlama-1.1B-Chat-v1.0.BF16 - Prompt Processing 2048 llamafile: TinyLlama-1.1B-Chat-v1.0.BF16 - Prompt Processing 1024 llamafile: TinyLlama-1.1B-Chat-v1.0.BF16 - Prompt Processing 512 llamafile: TinyLlama-1.1B-Chat-v1.0.BF16 - Prompt Processing 256 llamafile: Llama-3.2-3B-Instruct.Q6_K - Prompt Processing 2048 llamafile: Llama-3.2-3B-Instruct.Q6_K - Prompt Processing 1024 llamafile: Llama-3.2-3B-Instruct.Q6_K - Prompt Processing 512 llamafile: Llama-3.2-3B-Instruct.Q6_K - Prompt Processing 256 openvino-genai: Phi-3-mini-128k-instruct-int4-ov - CPU - Time Per Output Token openvino-genai: Phi-3-mini-128k-instruct-int4-ov - CPU - Time To First Token openvino-genai: Falcon-7b-instruct-int4-ov - CPU - Time Per Output Token openvino-genai: Falcon-7b-instruct-int4-ov - CPU - Time To First Token openvino-genai: TinyLlama-1.1B-Chat-v1.0 - CPU - Time Per Output Token openvino-genai: TinyLlama-1.1B-Chat-v1.0 - CPU - Time To First Token openvino-genai: Gemma-7b-int4-ov - CPU - Time Per Output Token openvino-genai: Gemma-7b-int4-ov - CPU - Time To First Token llamafile: wizardcoder-python-34b-v1.0.Q6_K - Text Generation 16 llamafile: mistral-7b-instruct-v0.2.Q5_K_M - Text Generation 16 llamafile: TinyLlama-1.1B-Chat-v1.0.BF16 - Text Generation 16 llamafile: Llama-3.2-3B-Instruct.Q6_K - Text Generation 16 dacapobench: PMD Source Code Analyzer couchdb: 500 - 1000 - 30 couchdb: 300 - 3000 - 30 dacapobench: Tradebeans relion: Basic - CPU palabos: 1000 palabos: 500 v6.11 v6.12 v6.13 7 Dec 36748330.54 192098 2724 4120 2071.98 522 233159294.14 8.498 117677 412967 12696519.85 4154651.85 2054.70 6.305 45376718.94 126877 2104 68.13 103.228 79773.37 2354 466.614 0.217 112.56 4621379 183693 159.90 43688 297.99 7421710.24 1053 923 2294 483952 21.66 770.47 3077 42.03 4058 4115352.49 0.167 339 15.19 6349 60.171 47.90 804.06 4783648 56.73 2093 801.19 718106876590 78.36 49.13 15.90 199.209 149.30 16.880 194.338 3782 561.91 47424.87 1.999 487834985450 5070 159.819 359931303710 333162995897 25.036 21.792 743 101.821 20.24 65.89 6791 12288 6144 3072 1536 32768 16384 8192 4096 32768 16384 8192 4096 32768 16384 8192 4096 15.18 22.66 17.63 30.39 12.76 15.86 20.36 30.39 14.98 64.87 152.07 108.22 1085 354.526 363.527 4999 176.673 481.489 529.232 55755486.32 183537 2259 4326 2298.65 551 251549880.01 8.927 112025 399417 13294800.59 4313954.94 2010.99 6.550 45398166.14 122181 2049 67.71 104.163 82674.98 2303 468.666 0.211 111.98 4732753 181192 159.60 43581 303.11 7297642.21 1038 937 2258 491186 21.49 765.59 3035 41.74 4071 4064048.14 0.166 335 15.08 6280 60.381 47.70 801.90 4815625 56.21 2083 799.99 721616676910 78.18 49.09 15.82 200.316 148.82 16.990 193.202 3755 565.10 47501.04 1.990 490511085203 5045 160.302 361564053870 334486696117 24.972 21.848 741 101.599 20.21 65.82 6797 12288 6144 3072 1536 32768 16384 8192 4096 32768 16384 8192 4096 32768 16384 8192 4096 15.19 22.73 17.80 30.00 12.79 15.89 20.37 30.27 15.01 64.80 152.89 107.95 1167 333.178 371.288 4968 179.868 655.184 775.110 53038526.00 158304 2651 3631 2385.59 493 246944948.46 9.082 110112 386983 13514659.67 4398887.77 2125.13 6.628 43186401.84 120703 2010 65.45 100.140 81582.58 2283 480.912 0.211 109.60 4735578 179909 156.82 42936 300.72 7400996.98 1055 938 2278 487743 21.35 774.01 3049 41.49 4019 4099640.31 0.165 338 15.02 6331 60.821 47.39 803.49 4829255 56.28 2102 807.27 719980763180 77.67 48.70 15.77 200.823 148.16 17.008 192.923 3763 565.14 47238.70 1.988 489486102253 5057 160.559 361324888880 334428157450 24.939 21.855 742 101.725 20.20 65.91 6793 12288 6144 3072 1536 32768 16384 8192 4096 32768 16384 8192 4096 32768 16384 8192 4096 15.17 22.38 17.77 30.18 12.88 16.05 20.54 30.48 14.90 64.15 147.98 102.21 1089 311.038 360.826 4957 172.482 509.126 764.024 OpenBenchmarking.org
Stress-NG Test: Context Switching OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.17.08 Test: Context Switching v6.11 v6.12 v6.13 7 Dec 12M 24M 36M 48M 60M SE +/- 247154.69, N = 13 SE +/- 64836.17, N = 3 SE +/- 159927.59, N = 3 36748330.54 55755486.32 53038526.00 1. (CXX) g++ options: -O2 -std=gnu99 -lc -lm
MariaDB Test: oltp_update_index - Threads: 128 OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 11.5 Test: oltp_update_index - Threads: 128 v6.11 v6.12 v6.13 7 Dec 40K 80K 120K 160K 200K SE +/- 123.94, N = 3 SE +/- 76.99, N = 3 SE +/- 112.30, N = 3 192098 183537 158304 1. (CXX) g++ options: -pie -fPIC -fstack-protector -O3 -lnuma -lpcre2-8 -lcrypt -laio -lz -lm -lssl -lcrypto -lpthread -ldl
DaCapo Benchmark Java Test: Spring Boot OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 23.11 Java Test: Spring Boot v6.11 v6.12 v6.13 7 Dec 600 1200 1800 2400 3000 SE +/- 17.89, N = 3 SE +/- 20.60, N = 3 SE +/- 26.77, N = 3 2724 2259 2651
DaCapo Benchmark Java Test: Apache Lucene Search Engine OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 23.11 Java Test: Apache Lucene Search Engine v6.11 v6.12 v6.13 7 Dec 900 1800 2700 3600 4500 SE +/- 43.31, N = 15 SE +/- 47.07, N = 3 SE +/- 42.57, N = 3 4120 4326 3631
Stress-NG Test: NUMA OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.17.08 Test: NUMA v6.11 v6.12 v6.13 7 Dec 500 1000 1500 2000 2500 SE +/- 6.78, N = 3 SE +/- 1.02, N = 3 SE +/- 10.25, N = 3 2071.98 2298.65 2385.59 1. (CXX) g++ options: -O2 -std=gnu99 -lc -lm
DaCapo Benchmark Java Test: Zxing 1D/2D Barcode Image Processing OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 23.11 Java Test: Zxing 1D/2D Barcode Image Processing v6.11 v6.12 v6.13 7 Dec 120 240 360 480 600 SE +/- 4.26, N = 3 SE +/- 4.38, N = 9 SE +/- 5.55, N = 3 522 551 493
Stress-NG Test: Semaphores OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.17.08 Test: Semaphores v6.11 v6.12 v6.13 7 Dec 50M 100M 150M 200M 250M SE +/- 2150155.18, N = 15 SE +/- 3074547.79, N = 3 SE +/- 1084579.30, N = 3 233159294.14 251549880.01 246944948.46 1. (CXX) g++ options: -O2 -std=gnu99 -lc -lm
PostgreSQL Scaling Factor: 100 - Clients: 1000 - Mode: Read Write - Average Latency OpenBenchmarking.org ms, Fewer Is Better PostgreSQL 17 Scaling Factor: 100 - Clients: 1000 - Mode: Read Write - Average Latency v6.11 v6.12 v6.13 7 Dec 3 6 9 12 15 SE +/- 0.020, N = 3 SE +/- 0.048, N = 3 SE +/- 0.042, N = 3 8.498 8.927 9.082 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpq -lpgcommon -lpgport -lm
PostgreSQL Scaling Factor: 100 - Clients: 1000 - Mode: Read Write OpenBenchmarking.org TPS, More Is Better PostgreSQL 17 Scaling Factor: 100 - Clients: 1000 - Mode: Read Write v6.11 v6.12 v6.13 7 Dec 30K 60K 90K 120K 150K SE +/- 274.82, N = 3 SE +/- 604.26, N = 3 SE +/- 515.38, N = 3 117677 112025 110112 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpq -lpgcommon -lpgport -lm
MariaDB Test: oltp_write_only - Threads: 128 OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 11.5 Test: oltp_write_only - Threads: 128 v6.11 v6.12 v6.13 7 Dec 90K 180K 270K 360K 450K SE +/- 286.92, N = 3 SE +/- 474.96, N = 3 SE +/- 154.44, N = 3 412967 399417 386983 1. (CXX) g++ options: -pie -fPIC -fstack-protector -O3 -lnuma -lpcre2-8 -lcrypt -laio -lz -lm -lssl -lcrypto -lpthread -ldl
Memcached Set To Get Ratio: 1:100 OpenBenchmarking.org Ops/sec, More Is Better Memcached 1.6.19 Set To Get Ratio: 1:100 v6.11 v6.12 v6.13 7 Dec 3M 6M 9M 12M 15M SE +/- 13756.93, N = 3 SE +/- 15384.68, N = 3 SE +/- 24482.38, N = 3 12696519.85 13294800.59 13514659.67 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
Stress-NG Test: Futex OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.17.08 Test: Futex v6.11 v6.12 v6.13 7 Dec 900K 1800K 2700K 3600K 4500K SE +/- 17548.84, N = 3 SE +/- 15124.18, N = 3 SE +/- 31680.82, N = 3 4154651.85 4313954.94 4398887.77 1. (CXX) g++ options: -O2 -std=gnu99 -lc -lm
Stress-NG Test: MEMFD OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.17.08 Test: MEMFD v6.11 v6.12 v6.13 7 Dec 500 1000 1500 2000 2500 SE +/- 4.79, N = 3 SE +/- 2.72, N = 3 SE +/- 3.27, N = 3 2054.70 2010.99 2125.13 1. (CXX) g++ options: -O2 -std=gnu99 -lc -lm
PostgreSQL Scaling Factor: 100 - Clients: 800 - Mode: Read Write - Average Latency OpenBenchmarking.org ms, Fewer Is Better PostgreSQL 17 Scaling Factor: 100 - Clients: 800 - Mode: Read Write - Average Latency v6.11 v6.12 v6.13 7 Dec 2 4 6 8 10 SE +/- 0.030, N = 3 SE +/- 0.082, N = 3 SE +/- 0.030, N = 3 6.305 6.550 6.628 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpq -lpgcommon -lpgport -lm
Stress-NG Test: Mutex OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.17.08 Test: Mutex v6.11 v6.12 v6.13 7 Dec 10M 20M 30M 40M 50M SE +/- 181454.22, N = 3 SE +/- 442253.04, N = 3 SE +/- 188378.96, N = 3 45376718.94 45398166.14 43186401.84 1. (CXX) g++ options: -O2 -std=gnu99 -lc -lm
PostgreSQL Scaling Factor: 100 - Clients: 800 - Mode: Read Write OpenBenchmarking.org TPS, More Is Better PostgreSQL 17 Scaling Factor: 100 - Clients: 800 - Mode: Read Write v6.11 v6.12 v6.13 7 Dec 30K 60K 90K 120K 150K SE +/- 594.42, N = 3 SE +/- 1549.82, N = 3 SE +/- 536.64, N = 3 126877 122181 120703 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpq -lpgcommon -lpgport -lm
DaCapo Benchmark Java Test: H2 Database Engine OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 23.11 Java Test: H2 Database Engine v6.11 v6.12 v6.13 7 Dec 500 1000 1500 2000 2500 SE +/- 28.09, N = 3 SE +/- 13.65, N = 3 SE +/- 18.21, N = 3 2104 2049 2010
Llamafile Model: mistral-7b-instruct-v0.2.Q5_K_M - Test: Text Generation 128 OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.8.16 Model: mistral-7b-instruct-v0.2.Q5_K_M - Test: Text Generation 128 v6.11 v6.12 v6.13 7 Dec 15 30 45 60 75 SE +/- 0.37, N = 3 SE +/- 0.47, N = 3 SE +/- 0.36, N = 3 68.13 67.71 65.45
Apache CouchDB Bulk Size: 300 - Inserts: 1000 - Rounds: 30 OpenBenchmarking.org Seconds, Fewer Is Better Apache CouchDB 3.4.1 Bulk Size: 300 - Inserts: 1000 - Rounds: 30 v6.11 v6.12 v6.13 7 Dec 20 40 60 80 100 SE +/- 1.11, N = 3 SE +/- 0.71, N = 3 SE +/- 0.87, N = 7 103.23 104.16 100.14 1. (CXX) g++ options: -flto -lstdc++ -shared -lei
Stress-NG Test: Mixed Scheduler OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.17.08 Test: Mixed Scheduler v6.11 v6.12 v6.13 7 Dec 20K 40K 60K 80K 100K SE +/- 161.54, N = 3 SE +/- 215.02, N = 3 SE +/- 282.83, N = 3 79773.37 82674.98 81582.58 1. (CXX) g++ options: -O2 -std=gnu99 -lc -lm
DaCapo Benchmark Java Test: Avrora AVR Simulation Framework OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 23.11 Java Test: Avrora AVR Simulation Framework v6.11 v6.12 v6.13 7 Dec 500 1000 1500 2000 2500 SE +/- 16.90, N = 3 SE +/- 15.56, N = 13 SE +/- 20.72, N = 15 2354 2303 2283
SVT-AV1 Encoder Mode: Preset 13 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2.3 Encoder Mode: Preset 13 - Input: Bosphorus 4K v6.11 v6.12 v6.13 7 Dec 100 200 300 400 500 SE +/- 3.40, N = 3 SE +/- 1.95, N = 3 SE +/- 0.44, N = 3 466.61 468.67 480.91 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
PostgreSQL Scaling Factor: 100 - Clients: 1000 - Mode: Read Only - Average Latency OpenBenchmarking.org ms, Fewer Is Better PostgreSQL 17 Scaling Factor: 100 - Clients: 1000 - Mode: Read Only - Average Latency v6.11 v6.12 v6.13 7 Dec 0.0488 0.0976 0.1464 0.1952 0.244 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 0.217 0.211 0.211 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpq -lpgcommon -lpgport -lm
Llamafile Model: Llama-3.2-3B-Instruct.Q6_K - Test: Text Generation 128 OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.8.16 Model: Llama-3.2-3B-Instruct.Q6_K - Test: Text Generation 128 v6.11 v6.12 v6.13 7 Dec 30 60 90 120 150 SE +/- 0.87, N = 3 SE +/- 1.04, N = 3 SE +/- 1.53, N = 3 112.56 111.98 109.60
PostgreSQL Scaling Factor: 100 - Clients: 1000 - Mode: Read Only OpenBenchmarking.org TPS, More Is Better PostgreSQL 17 Scaling Factor: 100 - Clients: 1000 - Mode: Read Only v6.11 v6.12 v6.13 7 Dec 1000K 2000K 3000K 4000K 5000K SE +/- 23812.59, N = 3 SE +/- 18553.35, N = 3 SE +/- 9098.55, N = 3 4621379 4732753 4735578 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpq -lpgcommon -lpgport -lm
MariaDB Test: oltp_read_write - Threads: 128 OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 11.5 Test: oltp_read_write - Threads: 128 v6.11 v6.12 v6.13 7 Dec 40K 80K 120K 160K 200K SE +/- 155.52, N = 3 SE +/- 60.34, N = 3 SE +/- 77.99, N = 3 183693 181192 179909 1. (CXX) g++ options: -pie -fPIC -fstack-protector -O3 -lnuma -lpcre2-8 -lcrypt -laio -lz -lm -lssl -lcrypto -lpthread -ldl
Llamafile Model: TinyLlama-1.1B-Chat-v1.0.BF16 - Test: Text Generation 128 OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.8.16 Model: TinyLlama-1.1B-Chat-v1.0.BF16 - Test: Text Generation 128 v6.11 v6.12 v6.13 7 Dec 40 80 120 160 200 SE +/- 1.11, N = 3 SE +/- 1.26, N = 3 SE +/- 1.45, N = 15 159.90 159.60 156.82
MariaDB Test: oltp_read_only - Threads: 128 OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 11.5 Test: oltp_read_only - Threads: 128 v6.11 v6.12 v6.13 7 Dec 9K 18K 27K 36K 45K SE +/- 14.78, N = 3 SE +/- 15.19, N = 3 SE +/- 18.45, N = 3 43688 43581 42936 1. (CXX) g++ options: -pie -fPIC -fstack-protector -O3 -lnuma -lpcre2-8 -lcrypt -laio -lz -lm -lssl -lcrypto -lpthread -ldl
Laghos Test: Triple Point Problem OpenBenchmarking.org Major Kernels Total Rate, More Is Better Laghos 3.1 Test: Triple Point Problem v6.11 v6.12 v6.13 7 Dec 70 140 210 280 350 SE +/- 3.08, N = 3 SE +/- 1.53, N = 3 SE +/- 3.23, N = 3 297.99 303.11 300.72 1. (CXX) g++ options: -O3 -std=c++11 -lmfem -lHYPRE -lmetis -lrt -lmpi_cxx -lmpi
Memcached Set To Get Ratio: 1:10 OpenBenchmarking.org Ops/sec, More Is Better Memcached 1.6.19 Set To Get Ratio: 1:10 v6.11 v6.12 v6.13 7 Dec 1.6M 3.2M 4.8M 6.4M 8M SE +/- 35717.40, N = 3 SE +/- 30399.87, N = 3 SE +/- 37597.18, N = 3 7421710.24 7297642.21 7400996.98 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
DaCapo Benchmark Java Test: Apache Tomcat OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 23.11 Java Test: Apache Tomcat v6.11 v6.12 v6.13 7 Dec 200 400 600 800 1000 SE +/- 7.81, N = 3 SE +/- 3.51, N = 3 SE +/- 2.00, N = 3 1053 1038 1055
DaCapo Benchmark Java Test: Batik SVG Toolkit OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 23.11 Java Test: Batik SVG Toolkit v6.11 v6.12 v6.13 7 Dec 200 400 600 800 1000 SE +/- 3.61, N = 3 SE +/- 4.00, N = 3 SE +/- 7.55, N = 3 923 937 938
DaCapo Benchmark Java Test: Apache Lucene Search Index OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 23.11 Java Test: Apache Lucene Search Index v6.11 v6.12 v6.13 7 Dec 500 1000 1500 2000 2500 SE +/- 16.64, N = 3 SE +/- 8.39, N = 3 SE +/- 25.32, N = 3 2294 2258 2278
Apache Cassandra Test: Writes OpenBenchmarking.org Op/s, More Is Better Apache Cassandra 5.0 Test: Writes v6.11 v6.12 v6.13 7 Dec 110K 220K 330K 440K 550K SE +/- 5140.53, N = 3 SE +/- 1282.82, N = 3 SE +/- 4648.88, N = 3 483952 491186 487743
Blender Blend File: Fishy Cat - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.3 Blend File: Fishy Cat - Compute: CPU-Only v6.11 v6.12 v6.13 7 Dec 5 10 15 20 25 SE +/- 0.12, N = 3 SE +/- 0.06, N = 3 SE +/- 0.06, N = 3 21.66 21.49 21.35
ClickHouse 100M Rows Hits Dataset, First Run / Cold Cache OpenBenchmarking.org Queries Per Minute, Geo Mean, More Is Better ClickHouse 22.12.3.5 100M Rows Hits Dataset, First Run / Cold Cache v6.11 v6.12 v6.13 7 Dec 200 400 600 800 1000 SE +/- 2.38, N = 3 SE +/- 5.98, N = 3 SE +/- 2.84, N = 3 770.47 765.59 774.01 MIN: 65.15 / MAX: 8571.43 MIN: 63.29 / MAX: 7500 MIN: 65.86 / MAX: 7500
DaCapo Benchmark Java Test: Tradesoap OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 23.11 Java Test: Tradesoap v6.11 v6.12 v6.13 7 Dec 700 1400 2100 2800 3500 SE +/- 24.92, N = 3 SE +/- 21.97, N = 15 SE +/- 31.71, N = 5 3077 3035 3049
Blender Blend File: Classroom - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.3 Blend File: Classroom - Compute: CPU-Only v6.11 v6.12 v6.13 7 Dec 10 20 30 40 50 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 42.03 41.74 41.49
DaCapo Benchmark Java Test: BioJava Biological Data Framework OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 23.11 Java Test: BioJava Biological Data Framework v6.11 v6.12 v6.13 7 Dec 900 1800 2700 3600 4500 SE +/- 11.89, N = 3 SE +/- 18.98, N = 3 SE +/- 22.00, N = 3 4058 4071 4019
Memcached Set To Get Ratio: 1:5 OpenBenchmarking.org Ops/sec, More Is Better Memcached 1.6.19 Set To Get Ratio: 1:5 v6.11 v6.12 v6.13 7 Dec 900K 1800K 2700K 3600K 4500K SE +/- 26467.49, N = 3 SE +/- 16085.40, N = 3 SE +/- 28528.57, N = 3 4115352.49 4064048.14 4099640.31 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
PostgreSQL Scaling Factor: 100 - Clients: 800 - Mode: Read Only - Average Latency OpenBenchmarking.org ms, Fewer Is Better PostgreSQL 17 Scaling Factor: 100 - Clients: 800 - Mode: Read Only - Average Latency v6.11 v6.12 v6.13 7 Dec 0.0376 0.0752 0.1128 0.1504 0.188 SE +/- 0.001, N = 3 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 0.167 0.166 0.165 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpq -lpgcommon -lpgport -lm
DaCapo Benchmark Java Test: FOP Print Formatter OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 23.11 Java Test: FOP Print Formatter v6.11 v6.12 v6.13 7 Dec 70 140 210 280 350 SE +/- 0.67, N = 3 SE +/- 2.08, N = 3 SE +/- 1.20, N = 3 339 335 338
Blender Blend File: BMW27 - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.3 Blend File: BMW27 - Compute: CPU-Only v6.11 v6.12 v6.13 7 Dec 4 8 12 16 20 SE +/- 0.00, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 15.19 15.08 15.02
DaCapo Benchmark Java Test: Eclipse OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 23.11 Java Test: Eclipse v6.11 v6.12 v6.13 7 Dec 1400 2800 4200 5600 7000 SE +/- 14.75, N = 3 SE +/- 33.39, N = 3 SE +/- 23.78, N = 3 6349 6280 6331
SVT-AV1 Encoder Mode: Preset 5 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2.3 Encoder Mode: Preset 5 - Input: Bosphorus 4K v6.11 v6.12 v6.13 7 Dec 14 28 42 56 70 SE +/- 0.07, N = 3 SE +/- 0.12, N = 3 SE +/- 0.08, N = 3 60.17 60.38 60.82 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
Blender Blend File: Pabellon Barcelona - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.3 Blend File: Pabellon Barcelona - Compute: CPU-Only v6.11 v6.12 v6.13 7 Dec 11 22 33 44 55 SE +/- 0.05, N = 3 SE +/- 0.08, N = 3 SE +/- 0.02, N = 3 47.90 47.70 47.39
ClickHouse 100M Rows Hits Dataset, Second Run OpenBenchmarking.org Queries Per Minute, Geo Mean, More Is Better ClickHouse 22.12.3.5 100M Rows Hits Dataset, Second Run v6.11 v6.12 v6.13 7 Dec 200 400 600 800 1000 SE +/- 4.73, N = 3 SE +/- 7.42, N = 3 SE +/- 8.62, N = 3 804.06 795.68 803.49 MIN: 65.15 / MAX: 10000 MIN: 65.29 / MAX: 7500 MIN: 66.08 / MAX: 10000
PostgreSQL Scaling Factor: 100 - Clients: 800 - Mode: Read Only OpenBenchmarking.org TPS, More Is Better PostgreSQL 17 Scaling Factor: 100 - Clients: 800 - Mode: Read Only v6.11 v6.12 v6.13 7 Dec 1000K 2000K 3000K 4000K 5000K SE +/- 11484.93, N = 3 SE +/- 9902.78, N = 3 SE +/- 8981.60, N = 3 4783648 4815625 4829255 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpq -lpgcommon -lpgport -lm
OpenVINO GenAI Model: Falcon-7b-instruct-int4-ov - Device: CPU OpenBenchmarking.org tokens/s, More Is Better OpenVINO GenAI 2024.5 Model: Falcon-7b-instruct-int4-ov - Device: CPU v6.11 v6.12 v6.13 7 Dec 13 26 39 52 65 SE +/- 0.63, N = 3 SE +/- 0.55, N = 6 SE +/- 0.54, N = 3 56.73 56.21 56.28
DaCapo Benchmark Java Test: GraphChi OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 23.11 Java Test: GraphChi v6.11 v6.12 v6.13 7 Dec 500 1000 1500 2000 2500 SE +/- 13.86, N = 3 SE +/- 5.51, N = 3 SE +/- 6.00, N = 3 2093 2083 2102
ClickHouse 100M Rows Hits Dataset, Third Run OpenBenchmarking.org Queries Per Minute, Geo Mean, More Is Better ClickHouse 22.12.3.5 100M Rows Hits Dataset, Third Run v6.11 v6.12 v6.13 7 Dec 200 400 600 800 1000 SE +/- 4.79, N = 3 SE +/- 7.60, N = 3 SE +/- 2.96, N = 3 801.19 799.99 807.27 MIN: 65.5 / MAX: 8571.43 MIN: 64.94 / MAX: 10000 MIN: 66.01 / MAX: 8571.43
OpenSSL Algorithm: ChaCha20 OpenBenchmarking.org byte/s, More Is Better OpenSSL Algorithm: ChaCha20 v6.11 v6.12 v6.13 7 Dec 160000M 320000M 480000M 640000M 800000M SE +/- 652717270.91, N = 3 SE +/- 1474303291.59, N = 3 SE +/- 626958911.68, N = 3 718106876590 721616676910 719980763180 1. OpenSSL 3.3.1 4 Jun 2024 (Library: OpenSSL 3.3.1 4 Jun 2024) - Additional Parameters: -engine qatengine -async_jobs 8
OpenVINO GenAI Model: TinyLlama-1.1B-Chat-v1.0 - Device: CPU OpenBenchmarking.org tokens/s, More Is Better OpenVINO GenAI 2024.5 Model: TinyLlama-1.1B-Chat-v1.0 - Device: CPU v6.11 v6.12 v6.13 7 Dec 20 40 60 80 100 SE +/- 0.57, N = 3 SE +/- 0.23, N = 3 SE +/- 0.28, N = 3 78.36 78.18 77.67
OpenVINO GenAI Model: Gemma-7b-int4-ov - Device: CPU OpenBenchmarking.org tokens/s, More Is Better OpenVINO GenAI 2024.5 Model: Gemma-7b-int4-ov - Device: CPU v6.11 v6.12 v6.13 7 Dec 11 22 33 44 55 SE +/- 0.15, N = 3 SE +/- 0.10, N = 3 SE +/- 0.27, N = 3 49.13 49.09 48.70
Llamafile Model: wizardcoder-python-34b-v1.0.Q6_K - Test: Text Generation 128 OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.8.16 Model: wizardcoder-python-34b-v1.0.Q6_K - Test: Text Generation 128 v6.11 v6.12 v6.13 7 Dec 4 8 12 16 20 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 15.90 15.82 15.77
SVT-AV1 Encoder Mode: Preset 8 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2.3 Encoder Mode: Preset 8 - Input: Bosphorus 4K v6.11 v6.12 v6.13 7 Dec 40 80 120 160 200 SE +/- 1.84, N = 3 SE +/- 1.77, N = 3 SE +/- 0.64, N = 3 199.21 200.32 200.82 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
Blender Blend File: Barbershop - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.3 Blend File: Barbershop - Compute: CPU-Only v6.11 v6.12 v6.13 7 Dec 30 60 90 120 150 SE +/- 0.08, N = 3 SE +/- 0.18, N = 3 SE +/- 0.23, N = 3 149.30 148.82 148.16
SVT-AV1 Encoder Mode: Preset 3 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2.3 Encoder Mode: Preset 3 - Input: Bosphorus 4K v6.11 v6.12 v6.13 7 Dec 4 8 12 16 20 SE +/- 0.08, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 16.88 16.99 17.01 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
Timed Linux Kernel Compilation Build: allmodconfig OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 6.8 Build: allmodconfig v6.11 v6.12 v6.13 7 Dec 40 80 120 160 200 SE +/- 0.21, N = 3 SE +/- 0.34, N = 3 SE +/- 0.25, N = 3 194.34 193.20 192.92
DaCapo Benchmark Java Test: Jython OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 23.11 Java Test: Jython v6.11 v6.12 v6.13 7 Dec 800 1600 2400 3200 4000 SE +/- 6.67, N = 3 SE +/- 5.33, N = 3 SE +/- 11.00, N = 3 3782 3755 3763
Laghos Test: Sedov Blast Wave, ube_922_hex.mesh OpenBenchmarking.org Major Kernels Total Rate, More Is Better Laghos 3.1 Test: Sedov Blast Wave, ube_922_hex.mesh v6.11 v6.12 v6.13 7 Dec 120 240 360 480 600 SE +/- 3.27, N = 3 SE +/- 2.47, N = 3 SE +/- 2.95, N = 3 561.91 565.10 565.14 1. (CXX) g++ options: -O3 -std=c++11 -lmfem -lHYPRE -lmetis -lrt -lmpi_cxx -lmpi
Stress-NG Test: Socket Activity OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.17.08 Test: Socket Activity v6.11 v6.12 v6.13 7 Dec 10K 20K 30K 40K 50K SE +/- 29.82, N = 3 SE +/- 68.22, N = 3 SE +/- 42.06, N = 3 47424.87 47501.04 47238.70 1. (CXX) g++ options: -O2 -std=gnu99 -lc -lm
Primesieve Length: 1e12 OpenBenchmarking.org Seconds, Fewer Is Better Primesieve 12.6 Length: 1e12 v6.11 v6.12 v6.13 7 Dec 0.4498 0.8996 1.3494 1.7992 2.249 SE +/- 0.003, N = 3 SE +/- 0.001, N = 3 SE +/- 0.002, N = 3 1.999 1.990 1.988 1. (CXX) g++ options: -O3
OpenSSL Algorithm: ChaCha20-Poly1305 OpenBenchmarking.org byte/s, More Is Better OpenSSL Algorithm: ChaCha20-Poly1305 v6.11 v6.12 v6.13 7 Dec 110000M 220000M 330000M 440000M 550000M SE +/- 271433364.08, N = 3 SE +/- 45557628.93, N = 3 SE +/- 158903312.00, N = 3 487834985450 489904220937 489486102253 1. OpenSSL 3.3.1 4 Jun 2024 (Library: OpenSSL 3.3.1 4 Jun 2024) - Additional Parameters: -engine qatengine -async_jobs 8
DaCapo Benchmark Java Test: Apache Kafka OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 23.11 Java Test: Apache Kafka v6.11 v6.12 v6.13 7 Dec 1100 2200 3300 4400 5500 SE +/- 7.33, N = 3 SE +/- 26.64, N = 3 SE +/- 3.53, N = 3 5070 5045 5057
Timed LLVM Compilation Build System: Unix Makefiles OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 16.0 Build System: Unix Makefiles v6.11 v6.12 v6.13 7 Dec 40 80 120 160 200 SE +/- 0.34, N = 3 SE +/- 0.33, N = 3 SE +/- 0.86, N = 3 159.82 160.30 160.56
OpenSSL Algorithm: AES-128-GCM OpenBenchmarking.org byte/s, More Is Better OpenSSL Algorithm: AES-128-GCM v6.11 v6.12 v6.13 7 Dec 80000M 160000M 240000M 320000M 400000M SE +/- 130266226.11, N = 3 SE +/- 210850977.29, N = 3 SE +/- 88875000.56, N = 3 359931303710 361513614630 361324888880 1. OpenSSL 3.3.1 4 Jun 2024 (Library: OpenSSL 3.3.1 4 Jun 2024) - Additional Parameters: -engine qatengine -async_jobs 8
OpenSSL Algorithm: AES-256-GCM OpenBenchmarking.org byte/s, More Is Better OpenSSL Algorithm: AES-256-GCM v6.11 v6.12 v6.13 7 Dec 70000M 140000M 210000M 280000M 350000M SE +/- 105448439.59, N = 3 SE +/- 189485712.03, N = 3 SE +/- 134252719.45, N = 3 333162995897 334279722140 334428157450 1. OpenSSL 3.3.1 4 Jun 2024 (Library: OpenSSL 3.3.1 4 Jun 2024) - Additional Parameters: -engine qatengine -async_jobs 8
Primesieve Length: 1e13 OpenBenchmarking.org Seconds, Fewer Is Better Primesieve 12.6 Length: 1e13 v6.11 v6.12 v6.13 7 Dec 6 12 18 24 30 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 25.04 24.97 24.94 1. (CXX) g++ options: -O3
Timed Linux Kernel Compilation Build: defconfig OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 6.8 Build: defconfig v6.11 v6.12 v6.13 7 Dec 5 10 15 20 25 SE +/- 0.25, N = 4 SE +/- 0.31, N = 3 SE +/- 0.26, N = 4 21.79 21.85 21.86
DaCapo Benchmark Java Test: Apache Xalan XSLT OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 23.11 Java Test: Apache Xalan XSLT v6.11 v6.12 v6.13 7 Dec 160 320 480 640 800 SE +/- 6.06, N = 3 SE +/- 1.76, N = 3 SE +/- 8.54, N = 3 743 741 742
Timed LLVM Compilation Build System: Ninja OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 16.0 Build System: Ninja v6.11 v6.12 v6.13 7 Dec 20 40 60 80 100 SE +/- 0.11, N = 3 SE +/- 0.06, N = 3 SE +/- 0.19, N = 3 101.82 101.60 101.73
Blender Blend File: Junkshop - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.3 Blend File: Junkshop - Compute: CPU-Only v6.11 v6.12 v6.13 7 Dec 5 10 15 20 25 SE +/- 0.04, N = 3 SE +/- 0.06, N = 3 SE +/- 0.06, N = 3 20.24 20.21 20.20
OpenVINO GenAI Model: Phi-3-mini-128k-instruct-int4-ov - Device: CPU OpenBenchmarking.org tokens/s, More Is Better OpenVINO GenAI 2024.5 Model: Phi-3-mini-128k-instruct-int4-ov - Device: CPU v6.11 v6.12 v6.13 7 Dec 15 30 45 60 75 SE +/- 0.41, N = 3 SE +/- 0.10, N = 3 SE +/- 0.79, N = 3 65.89 65.82 65.91
DaCapo Benchmark Java Test: jMonkeyEngine OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 23.11 Java Test: jMonkeyEngine v6.11 v6.12 v6.13 7 Dec 1500 3000 4500 6000 7500 SE +/- 1.86, N = 3 SE +/- 4.84, N = 3 SE +/- 1.86, N = 3 6791 6797 6793
Llamafile Model: wizardcoder-python-34b-v1.0.Q6_K - Test: Prompt Processing 2048 OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.8.16 Model: wizardcoder-python-34b-v1.0.Q6_K - Test: Prompt Processing 2048 v6.11 v6.12 v6.13 7 Dec 3K 6K 9K 12K 15K SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 12288 12288 12288
Llamafile Model: wizardcoder-python-34b-v1.0.Q6_K - Test: Prompt Processing 1024 OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.8.16 Model: wizardcoder-python-34b-v1.0.Q6_K - Test: Prompt Processing 1024 v6.11 v6.12 v6.13 7 Dec 1300 2600 3900 5200 6500 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 6144 6144 6144
Llamafile Model: wizardcoder-python-34b-v1.0.Q6_K - Test: Prompt Processing 512 OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.8.16 Model: wizardcoder-python-34b-v1.0.Q6_K - Test: Prompt Processing 512 v6.11 v6.12 v6.13 7 Dec 700 1400 2100 2800 3500 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 3072 3072 3072
Llamafile Model: wizardcoder-python-34b-v1.0.Q6_K - Test: Prompt Processing 256 OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.8.16 Model: wizardcoder-python-34b-v1.0.Q6_K - Test: Prompt Processing 256 v6.11 v6.12 v6.13 7 Dec 300 600 900 1200 1500 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1536 1536 1536
Llamafile Model: mistral-7b-instruct-v0.2.Q5_K_M - Test: Prompt Processing 2048 OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.8.16 Model: mistral-7b-instruct-v0.2.Q5_K_M - Test: Prompt Processing 2048 v6.11 v6.12 v6.13 7 Dec 7K 14K 21K 28K 35K SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 32768 32768 32768
Llamafile Model: mistral-7b-instruct-v0.2.Q5_K_M - Test: Prompt Processing 1024 OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.8.16 Model: mistral-7b-instruct-v0.2.Q5_K_M - Test: Prompt Processing 1024 v6.11 v6.12 v6.13 7 Dec 4K 8K 12K 16K 20K SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 16384 16384 16384
Llamafile Model: mistral-7b-instruct-v0.2.Q5_K_M - Test: Prompt Processing 512 OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.8.16 Model: mistral-7b-instruct-v0.2.Q5_K_M - Test: Prompt Processing 512 v6.11 v6.12 v6.13 7 Dec 2K 4K 6K 8K 10K SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 8192 8192 8192
Llamafile Model: mistral-7b-instruct-v0.2.Q5_K_M - Test: Prompt Processing 256 OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.8.16 Model: mistral-7b-instruct-v0.2.Q5_K_M - Test: Prompt Processing 256 v6.11 v6.12 v6.13 7 Dec 900 1800 2700 3600 4500 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 4096 4096 4096
Llamafile Model: TinyLlama-1.1B-Chat-v1.0.BF16 - Test: Prompt Processing 2048 OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.8.16 Model: TinyLlama-1.1B-Chat-v1.0.BF16 - Test: Prompt Processing 2048 v6.11 v6.12 v6.13 7 Dec 7K 14K 21K 28K 35K SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 32768 32768 32768
Llamafile Model: TinyLlama-1.1B-Chat-v1.0.BF16 - Test: Prompt Processing 1024 OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.8.16 Model: TinyLlama-1.1B-Chat-v1.0.BF16 - Test: Prompt Processing 1024 v6.11 v6.12 v6.13 7 Dec 4K 8K 12K 16K 20K SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 16384 16384 16384
Llamafile Model: TinyLlama-1.1B-Chat-v1.0.BF16 - Test: Prompt Processing 512 OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.8.16 Model: TinyLlama-1.1B-Chat-v1.0.BF16 - Test: Prompt Processing 512 v6.11 v6.12 v6.13 7 Dec 2K 4K 6K 8K 10K SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 8192 8192 8192
Llamafile Model: TinyLlama-1.1B-Chat-v1.0.BF16 - Test: Prompt Processing 256 OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.8.16 Model: TinyLlama-1.1B-Chat-v1.0.BF16 - Test: Prompt Processing 256 v6.11 v6.12 v6.13 7 Dec 900 1800 2700 3600 4500 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 4096 4096 4096
Llamafile Model: Llama-3.2-3B-Instruct.Q6_K - Test: Prompt Processing 2048 OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.8.16 Model: Llama-3.2-3B-Instruct.Q6_K - Test: Prompt Processing 2048 v6.11 v6.12 v6.13 7 Dec 7K 14K 21K 28K 35K SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 32768 32768 32768
Llamafile Model: Llama-3.2-3B-Instruct.Q6_K - Test: Prompt Processing 1024 OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.8.16 Model: Llama-3.2-3B-Instruct.Q6_K - Test: Prompt Processing 1024 v6.11 v6.12 v6.13 7 Dec 4K 8K 12K 16K 20K SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 16384 16384 16384
Llamafile Model: Llama-3.2-3B-Instruct.Q6_K - Test: Prompt Processing 512 OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.8.16 Model: Llama-3.2-3B-Instruct.Q6_K - Test: Prompt Processing 512 v6.11 v6.12 v6.13 7 Dec 2K 4K 6K 8K 10K SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 8192 8192 8192
Llamafile Model: Llama-3.2-3B-Instruct.Q6_K - Test: Prompt Processing 256 OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.8.16 Model: Llama-3.2-3B-Instruct.Q6_K - Test: Prompt Processing 256 v6.11 v6.12 v6.13 7 Dec 900 1800 2700 3600 4500 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 4096 4096 4096
OpenVINO GenAI Model: Phi-3-mini-128k-instruct-int4-ov - Device: CPU - Time Per Output Token OpenBenchmarking.org ms, Fewer Is Better OpenVINO GenAI 2024.5 Model: Phi-3-mini-128k-instruct-int4-ov - Device: CPU - Time Per Output Token v6.11 v6.12 v6.13 7 Dec 4 8 12 16 20 SE +/- 0.10, N = 3 SE +/- 0.02, N = 3 SE +/- 0.18, N = 3 15.18 15.19 15.17
OpenVINO GenAI Model: Phi-3-mini-128k-instruct-int4-ov - Device: CPU - Time To First Token OpenBenchmarking.org ms, Fewer Is Better OpenVINO GenAI 2024.5 Model: Phi-3-mini-128k-instruct-int4-ov - Device: CPU - Time To First Token v6.11 v6.12 v6.13 7 Dec 5 10 15 20 25 SE +/- 0.24, N = 3 SE +/- 0.37, N = 3 SE +/- 0.43, N = 3 22.66 22.73 22.38
OpenVINO GenAI Model: Falcon-7b-instruct-int4-ov - Device: CPU - Time Per Output Token OpenBenchmarking.org ms, Fewer Is Better OpenVINO GenAI 2024.5 Model: Falcon-7b-instruct-int4-ov - Device: CPU - Time Per Output Token v6.11 v6.12 v6.13 7 Dec 4 8 12 16 20 SE +/- 0.19, N = 3 SE +/- 0.17, N = 6 SE +/- 0.17, N = 3 17.63 17.80 17.77
OpenVINO GenAI Model: Falcon-7b-instruct-int4-ov - Device: CPU - Time To First Token OpenBenchmarking.org ms, Fewer Is Better OpenVINO GenAI 2024.5 Model: Falcon-7b-instruct-int4-ov - Device: CPU - Time To First Token v6.11 v6.12 v6.13 7 Dec 7 14 21 28 35 SE +/- 0.27, N = 3 SE +/- 0.26, N = 6 SE +/- 0.44, N = 3 30.39 30.00 30.18
OpenVINO GenAI Model: TinyLlama-1.1B-Chat-v1.0 - Device: CPU - Time Per Output Token OpenBenchmarking.org ms, Fewer Is Better OpenVINO GenAI 2024.5 Model: TinyLlama-1.1B-Chat-v1.0 - Device: CPU - Time Per Output Token v6.11 v6.12 v6.13 7 Dec 3 6 9 12 15 SE +/- 0.09, N = 3 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 12.76 12.79 12.88
OpenVINO GenAI Model: TinyLlama-1.1B-Chat-v1.0 - Device: CPU - Time To First Token OpenBenchmarking.org ms, Fewer Is Better OpenVINO GenAI 2024.5 Model: TinyLlama-1.1B-Chat-v1.0 - Device: CPU - Time To First Token v6.11 v6.12 v6.13 7 Dec 4 8 12 16 20 SE +/- 0.19, N = 3 SE +/- 0.05, N = 3 SE +/- 0.08, N = 3 15.86 15.89 16.05
OpenVINO GenAI Model: Gemma-7b-int4-ov - Device: CPU - Time Per Output Token OpenBenchmarking.org ms, Fewer Is Better OpenVINO GenAI 2024.5 Model: Gemma-7b-int4-ov - Device: CPU - Time Per Output Token v6.11 v6.12 v6.13 7 Dec 5 10 15 20 25 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 SE +/- 0.11, N = 3 20.36 20.37 20.54
OpenVINO GenAI Model: Gemma-7b-int4-ov - Device: CPU - Time To First Token OpenBenchmarking.org ms, Fewer Is Better OpenVINO GenAI 2024.5 Model: Gemma-7b-int4-ov - Device: CPU - Time To First Token v6.11 v6.12 v6.13 7 Dec 7 14 21 28 35 SE +/- 0.08, N = 3 SE +/- 0.04, N = 3 SE +/- 0.18, N = 3 30.39 30.27 30.48
Llamafile Model: wizardcoder-python-34b-v1.0.Q6_K - Test: Text Generation 16 OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.8.16 Model: wizardcoder-python-34b-v1.0.Q6_K - Test: Text Generation 16 v6.11 v6.12 v6.13 7 Dec 4 8 12 16 20 SE +/- 0.74, N = 12 SE +/- 0.74, N = 12 SE +/- 0.73, N = 12 14.98 15.01 14.90
Llamafile Model: mistral-7b-instruct-v0.2.Q5_K_M - Test: Text Generation 16 OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.8.16 Model: mistral-7b-instruct-v0.2.Q5_K_M - Test: Text Generation 16 v6.11 v6.12 v6.13 7 Dec 14 28 42 56 70 SE +/- 1.73, N = 12 SE +/- 1.75, N = 12 SE +/- 1.70, N = 12 64.87 64.80 64.15
Llamafile Model: TinyLlama-1.1B-Chat-v1.0.BF16 - Test: Text Generation 16 OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.8.16 Model: TinyLlama-1.1B-Chat-v1.0.BF16 - Test: Text Generation 16 v6.11 v6.12 v6.13 7 Dec 30 60 90 120 150 SE +/- 4.56, N = 12 SE +/- 4.63, N = 12 SE +/- 4.35, N = 12 152.07 152.89 147.98
Llamafile Model: Llama-3.2-3B-Instruct.Q6_K - Test: Text Generation 16 OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.8.16 Model: Llama-3.2-3B-Instruct.Q6_K - Test: Text Generation 16 v6.11 v6.12 v6.13 7 Dec 20 40 60 80 100 SE +/- 2.48, N = 12 SE +/- 2.55, N = 12 SE +/- 2.52, N = 14 108.22 107.95 102.21
DaCapo Benchmark Java Test: PMD Source Code Analyzer OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 23.11 Java Test: PMD Source Code Analyzer v6.11 v6.12 v6.13 7 Dec 300 600 900 1200 1500 SE +/- 5.90, N = 3 SE +/- 23.97, N = 15 SE +/- 18.53, N = 15 1085 1167 1089
Apache CouchDB Bulk Size: 500 - Inserts: 1000 - Rounds: 30 OpenBenchmarking.org Seconds, Fewer Is Better Apache CouchDB 3.4.1 Bulk Size: 500 - Inserts: 1000 - Rounds: 30 v6.11 v6.12 v6.13 7 Dec 80 160 240 320 400 SE +/- 62.71, N = 8 SE +/- 36.20, N = 9 SE +/- 29.09, N = 7 354.53 333.18 311.04 1. (CXX) g++ options: -flto -lstdc++ -shared -lei
Apache CouchDB Bulk Size: 300 - Inserts: 3000 - Rounds: 30 OpenBenchmarking.org Seconds, Fewer Is Better Apache CouchDB 3.4.1 Bulk Size: 300 - Inserts: 3000 - Rounds: 30 v6.11 v6.12 v6.13 7 Dec 80 160 240 320 400 SE +/- 6.26, N = 9 SE +/- 12.48, N = 9 SE +/- 3.45, N = 9 363.53 371.29 360.83 1. (CXX) g++ options: -flto -lstdc++ -shared -lei
DaCapo Benchmark Java Test: Tradebeans OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 23.11 Java Test: Tradebeans v6.11 v6.12 v6.13 7 Dec 1100 2200 3300 4400 5500 SE +/- 48.79, N = 15 SE +/- 76.59, N = 15 SE +/- 100.21, N = 12 4999 4968 4957
RELION Test: Basic - Device: CPU OpenBenchmarking.org Seconds, Fewer Is Better RELION 5.0 Test: Basic - Device: CPU v6.11 v6.12 v6.13 7 Dec 40 80 120 160 200 SE +/- 1.95, N = 3 SE +/- 0.79, N = 3 SE +/- 3.31, N = 12 176.67 179.87 172.48 1. (CXX) g++ options: -fPIC -std=c++14 -fopenmp -O3 -rdynamic -ldl -ltiff -lfftw3f -lfftw3 -lpng -ljpeg -lmpi_cxx -lmpi
Palabos Grid Size: 1000 OpenBenchmarking.org Mega Site Updates Per Second, More Is Better Palabos 2.3 Grid Size: 1000 v6.11 v6.12 v6.13 7 Dec 140 280 420 560 700 SE +/- 3.07, N = 15 SE +/- 36.69, N = 12 SE +/- 21.98, N = 15 481.49 655.18 509.13 1. (CXX) g++ options: -std=c++17 -pedantic -O3 -rdynamic -lcrypto -lcurl -lsz -lz -ldl -lm
Palabos Grid Size: 500 OpenBenchmarking.org Mega Site Updates Per Second, More Is Better Palabos 2.3 Grid Size: 500 v6.11 v6.12 v6.13 7 Dec 200 400 600 800 1000 SE +/- 40.67, N = 12 SE +/- 1.19, N = 3 SE +/- 5.28, N = 3 529.23 775.11 764.02 1. (CXX) g++ options: -std=c++17 -pedantic -O3 -rdynamic -lcrypto -lcurl -lsz -lz -ldl -lm
Phoronix Test Suite v10.8.5