Tests
Suites
Latest Results
Search
Register
Login
Popular Tests
Flexible IO Tester
Llama.cpp
Timed Linux Kernel Compilation
Blender
PostgreSQL
Hashcat
Newest Tests
OpenVINO GenAI
Rustls
LiteRT
Recently Updated Tests
QMCPACK
Liquid-DSP
Llama.cpp
srsRAN Project
NAMD CUDA
Stress-NG
New & Recently Updated Tests
Recently Updated Suites
Machine Learning
Server Motherboard
HPC - High Performance Computing
New & Recently Updated Suites
Component Benchmarks
CPUs / Processors
GPUs / Graphics
OpenGL
Disks / Storage
Motherboards
File-Systems
Operating Systems
OpenBenchmarking.org
Corporate / Organization Info
Bug Reports / Feature Requests
Llama.cpp 2.1.0
pts/llama-cpp-2.1.0
- 29 December 2024 -
Update against Llama.cpp b4397 upstream.
downloads.xml
<?xml version="1.0"?> <!--Phoronix Test Suite v10.8.5--> <PhoronixTestSuite> <Downloads> <Package> <URL>https://github.com/ggerganov/llama.cpp/archive/refs/tags/b4397.tar.gz</URL> <MD5>3c471c3828bd4bdb054fba7c9427d3c5</MD5> <SHA256>51e5ac9b5c18756cf46ebc0659dcd57542800dc6582bbdf737dee064aaebfa34</SHA256> <FileName>llama.cpp-b4397.tar.gz</FileName> <FileSize>20603283</FileSize> </Package> <Package> <URL>https://huggingface.co/lmstudio-community/Llama-3.1-Tulu-3-8B-GGUF/resolve/7033c16b4f79f8708a27d80bf2ae0c6537253d1b/Llama-3.1-Tulu-3-8B-Q8_0.gguf?download=true</URL> <MD5>68a32ec44ea01a92c116a0b6fb83eae8</MD5> <SHA256>388db24ac65abaf1cd9dc9a0fc8d5aaebde5df908048b89c8cf3c2cec92562ef</SHA256> <FileName>Llama-3.1-Tulu-3-8B-Q8_0.gguf</FileName> <FileSize>8540841632</FileSize> <Optional>TRUE</Optional> </Package> <Package> <URL>https://huggingface.co/lmstudio-community/granite-3.0-3b-a800m-instruct-GGUF/resolve/46f75d55362bc1d5152541ec9579d38381ad7a59/granite-3.0-3b-a800m-instruct-Q8_0.gguf?download=true</URL> <MD5>faec97f7c662b271864eef04879d35fa</MD5> <SHA256>441d5c1195113695a10afb0ce7c105e9fc3c8c6d12960dab1716474326668b41</SHA256> <FileName>granite-3.0-3b-a800m-instruct-Q8_0.gguf</FileName> <FileSize>3592999808</FileSize> <Optional>TRUE</Optional> </Package> <Package> <URL>https://huggingface.co/lmstudio-community/Mistral-7B-Instruct-v0.3-GGUF/resolve/29a785419661afc70b5cd91b5023a835b0092281/Mistral-7B-Instruct-v0.3-Q8_0.gguf?download=true</URL> <MD5>5bc8f99351f114d3a323b9f7d1da846a</MD5> <SHA256>404857e776114baada71a08ebd3bba79d721ec7fca99705e7e7b892ae8bc583f</SHA256> <FileName>Mistral-7B-Instruct-v0.3-Q8_0.gguf</FileName> <FileSize>7702565088</FileSize> <Optional>TRUE</Optional> </Package> </Downloads> </PhoronixTestSuite>
install.sh
#!/bin/bash tar -xf llama.cpp-b4397.tar.gz rm -rf llama.cpp-BLAS cp -va llama.cpp-b4397 llama.cpp-BLAS cd llama.cpp-BLAS cmake -B build -DGGML_BLAS=ON -DGGML_BLAS_VENDOR=OpenBLAS cmake --build build --config Release -j $NUM_CPU_CORES echo $? > ~/install-exit-status rm -rf llama.cpp-b4397 echo "#!/bin/sh LLAMA_BENCH_ARGS=\`echo \"\$@\" | sed \"s/\$1/ /g\"\` cd llama.cpp-\$1 ./build/bin/llama-bench -t \$NUM_CPU_PHYSICAL_CORES \$LLAMA_BENCH_ARGS > \$LOG_FILE 2>&1 echo \$? > ~/test-exit-status" > ~/llama-cpp chmod +x ~/llama-cpp
results-definition.xml
<?xml version="1.0"?> <!--Phoronix Test Suite v10.8.5--> <PhoronixTestSuite> <ResultsParser> <OutputTemplate>| llama 8B Q8_0 | 7.95 GiB | 8.03 B | BLAS | 128 | tg128 | #_RESULT_# ± 0.29 |</OutputTemplate> <LineHint>±</LineHint> <ResultBeforeString>±</ResultBeforeString> </ResultsParser> </PhoronixTestSuite>
test-definition.xml
<?xml version="1.0"?> <!--Phoronix Test Suite v10.8.5--> <PhoronixTestSuite> <TestInformation> <Title>Llama.cpp</Title> <AppVersion>b4397</AppVersion> <Description>Llama.cpp is a port of Facebook's LLaMA model in C/C++ developed by Georgi Gerganov. Llama.cpp allows the inference of LLaMA and other supported models in C/C++. For CPU inference Llama.cpp supports AVX2/AVX-512, ARM NEON, and other modern ISAs along with features like OpenBLAS usage.</Description> <ResultScale>Tokens Per Second</ResultScale> <Proportion>HIB</Proportion> <TimesToRun>3</TimesToRun> </TestInformation> <TestProfile> <Version>2.1.0</Version> <SupportedPlatforms>Linux</SupportedPlatforms> <SoftwareType>Utility</SoftwareType> <TestType>System</TestType> <License>Free</License> <ExternalDependencies>build-utilities, blas-development, cmake</ExternalDependencies> <InstallRequiresInternet>TRUE</InstallRequiresInternet> <EnvironmentSize>58700</EnvironmentSize> <ProjectURL>https://github.com/ggerganov/llama.cpp/</ProjectURL> <RepositoryURL>https://github.com/ggerganov/llama.cpp</RepositoryURL> <Maintainer>Michael Larabel</Maintainer> <SystemDependencies>pkgconf</SystemDependencies> </TestProfile> <TestSettings> <Option> <DisplayName>Backend</DisplayName> <Identifier>backend</Identifier> <Menu> <Entry> <Name>CPU BLAS</Name> <Value>BLAS</Value> </Entry> </Menu> </Option> <Option> <DisplayName>Model</DisplayName> <Identifier>model</Identifier> <ArgumentPrefix>-m ../</ArgumentPrefix> <Menu> <Entry> <Name>Llama-3.1-Tulu-3-8B-Q8_0</Name> <Value>Llama-3.1-Tulu-3-8B-Q8_0.gguf</Value> </Entry> <Entry> <Name>granite-3.0-3b-a800m-instruct-Q8_0</Name> <Value>granite-3.0-3b-a800m-instruct-Q8_0.gguf</Value> </Entry> <Entry> <Name>Mistral-7B-Instruct-v0.3-Q8_0</Name> <Value>Mistral-7B-Instruct-v0.3-Q8_0.gguf</Value> </Entry> </Menu> </Option> <Option> <DisplayName>Test</DisplayName> <Identifier>test</Identifier> <Menu> <Entry> <Name>Text Generation 128</Name> <Value>-n 128 -p 0</Value> </Entry> <Entry> <Name>Prompt Processing 512</Name> <Value>-n 0 -p 512</Value> </Entry> <Entry> <Name>Prompt Processing 1024</Name> <Value>-n 0 -p 1024</Value> </Entry> <Entry> <Name>Prompt Processing 2048</Name> <Value>-n 0 -p 2048</Value> </Entry> </Menu> </Option> </TestSettings> </PhoronixTestSuite>