Llama.cpp 1.0.0

pts/llama-cpp-1.0.0 - 10 January 2024 - Initial commit of llama.cpp CPU benchmark.

downloads.xml

<?xml version="1.0"?>
<!--Phoronix Test Suite v10.8.4-->
<PhoronixTestSuite>
  <Downloads>
    <Package>
      <URL>https://github.com/ggerganov/llama.cpp/archive/refs/tags/b1808.tar.gz</URL>
      <MD5>3a659d857520ca4d95f81f1fd03c4bca</MD5>
      <SHA256>f17b18178d7174d9659df74e5832285ccf5dad0c759627500f35aab0aa5ca14f</SHA256>
      <FileName>llama.cpp-b1808.tar.gz</FileName>
      <FileSize>7896552</FileSize>
    </Package>
    <Package>
      <URL>https://huggingface.co/TheBloke/Llama-2-13B-GGUF/resolve/50b3e202c3df49b96551490582cd234472c5eb23/llama-2-13b.Q4_0.gguf</URL>
      <MD5>5bd7db6ab9d3b2c312e46738e89090b3</MD5>
      <SHA256>283fc12bcea5638fc82bcd038f0d19eaf0ebdc7fec9e4536d3e8063c1cc84548</SHA256>
      <FileName>llama-2-13b.Q4_0.gguf</FileName>
      <FileSize>7365834624</FileSize>
      <Optional>TRUE</Optional>
    </Package>
    <Package>
      <URL>https://huggingface.co/TheBloke/Llama-2-70B-Chat-GGUF/resolve/96fbf8c3617a084d06d6947e98e0194aa818e5e8/llama-2-70b-chat.Q5_0.gguf</URL>
      <MD5>49dd1daed8811dc92c60952954d1e4a9</MD5>
      <SHA256>65a8fda53f4c7c6470fd3a4b8e798b80e8b1b7d6be34cd6b521a9a251eeb9c1a</SHA256>
      <FileName>llama-2-70b-chat.Q5_0.gguf</FileName>
      <FileSize>47461397408</FileSize>
      <Optional>TRUE</Optional>
    </Package>
    <Package>
      <URL>https://huggingface.co/TheBloke/Llama-2-7B-GGUF/resolve/b4e04e128f421c93a5f1e34ac4d7ca9b0af47b80/llama-2-7b.Q4_0.gguf</URL>
      <MD5>4344ea31374d6fd63023ee035c9f95ae</MD5>
      <SHA256>78b8f9777dd620ad29cd2cffb6653b17fa8a5b1fddc1b8821180d60eedd24d48</SHA256>
      <FileName>llama-2-7b.Q4_0.gguf</FileName>
      <FileSize>3825807040</FileSize>
      <Optional>TRUE</Optional>
    </Package>
  </Downloads>
</PhoronixTestSuite>

install.sh

#!/bin/bash
tar -xf llama.cpp-b1808.tar.gz
cd llama.cpp-b1808
make -j LLAMA_OPENBLAS=1 
echo $? > ~/install-exit-status
echo "#!/bin/sh
cd llama.cpp-b1808
./main \$@ -p \"Building a website can be done in 10 simple steps:\" -n 512 -e -t \$NUM_CPU_PHYSICAL_CORES > \$LOG_FILE 2>&1
echo \$? > ~/test-exit-status" > ~/llama-cpp
chmod +x ~/llama-cpp

results-definition.xml

<?xml version="1.0"?>
<!--Phoronix Test Suite v10.8.4-->
<PhoronixTestSuite>
  <ResultsParser>
    <OutputTemplate>llama_print_timings:        eval time =   18329.86 ms /   399 runs   (   45.94 ms per token,    #_RESULT_# tokens per second)</OutputTemplate>
    <ResultBeforeString>tokens</ResultBeforeString>
  </ResultsParser>
</PhoronixTestSuite>

test-definition.xml

<?xml version="1.0"?>
<!--Phoronix Test Suite v10.8.4-->
<PhoronixTestSuite>
  <TestInformation>
    <Title>Llama.cpp</Title>
    <AppVersion>b1808</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>1.0.0</Version>
    <SupportedPlatforms>Linux</SupportedPlatforms>
    <SoftwareType>Utility</SoftwareType>
    <TestType>System</TestType>
    <License>Free</License>
    <ExternalDependencies>build-utilities, blas-development</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>Model</DisplayName>
      <Identifier>model</Identifier>
      <ArgumentPrefix>-m ../</ArgumentPrefix>
      <Menu>
        <Entry>
          <Name>llama-2-7b.Q4_0.gguf</Name>
          <Value>llama-2-7b.Q4_0.gguf</Value>
        </Entry>
        <Entry>
          <Name>llama-2-13b.Q4_0.gguf</Name>
          <Value>llama-2-13b.Q4_0.gguf</Value>
        </Entry>
        <Entry>
          <Name>llama-2-70b-chat.Q5_0.gguf</Name>
          <Value>llama-2-70b-chat.Q5_0.gguf</Value>
        </Entry>
      </Menu>
    </Option>
  </TestSettings>
</PhoronixTestSuite>