Tests
Suites
Latest Results
Search
Register
Login
Popular Tests
Timed Linux Kernel Compilation
SVT-AV1
7-Zip Compression
Stockfish
FFmpeg
x265
Newest Tests
Rustls
LiteRT
WarpX
Epoch
Valkey
Whisperfile
Recently Updated Tests
Mobile Neural Network
ACES DGEMM
NWChem
SuperTuxKart
ASTC Encoder
SVT-AV1
New & Recently Updated Tests
Recently Updated Suites
Database Test Suite
Machine Learning
Steam
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
Llamafile 1.0.0
pts/llamafile-1.0.0
- 19 January 2024 -
Initial commit.
downloads.xml
<?xml version="1.0"?> <!--Phoronix Test Suite v10.8.4--> <PhoronixTestSuite> <Downloads> <Package> <URL>https://huggingface.co/jartine/Mistral-7B-Instruct-v0.2-llamafile/resolve/afe7a20630a30479686529d8ff878772cc04ddde/mistral-7b-instruct-v0.2.Q8_0.llamafile?download=true</URL> <MD5>1b141d44fbabc994c1c734c01e62b4e7</MD5> <SHA256>ea074763987d7e550b386d0b0e92ab441c05bb48fa0fed91db78c69a9a8c0adf</SHA256> <FileName>mistral-7b-instruct-v0.2.Q8_0.llamafile</FileName> <FileSize>7726305670</FileSize> <Optional>TRUE</Optional> </Package> <Package> <URL>https://huggingface.co/jartine/llava-v1.5-7B-GGUF/resolve/675a459780ff40b1c406f9760956ff5e2248d0d1/llava-v1.5-7b-q4.llamafile?download=true</URL> <MD5>282ea2378a02491f8e0ac1df280a5561</MD5> <SHA256>9c37a9a8e3f067dea8c028db9525b399fc53b267667ed9c2a60155b1aa7589de</SHA256> <FileName>llava-v1.5-7b-q4.llamafile</FileName> <FileSize>4288952500</FileSize> <Optional>TRUE</Optional> </Package> <Package> <URL>https://huggingface.co/jartine/WizardCoder-Python-34B-V1.0-llamafile/resolve/780ff2983b49772ef9482e60e11cc4de5fbd0d88/wizardcoder-python-34b-v1.0.Q6_K.llamafile?download=true</URL> <MD5>3c3f2bb47522892b2dea8963c44b17de</MD5> <SHA256>a5f96429b3831f5d665ce7e01be2b9037572ce527dde6159adf2c2398d3cc827</SHA256> <FileName>wizardcoder-python-34b-v1.0.Q6_K.llamafile</FileName> <FileSize>27714326924</FileSize> <Optional>TRUE</Optional> </Package> <Package> <URL>http://www.phoronix-test-suite.com/benchmark-files/pts-sample-photos-2.tar.bz2, http://www.phoronix.net/downloads/phoronix-test-suite/benchmark-files/pts-sample-photos-2.tar.bz2</URL> <MD5>7ff2870cf3fa2299ea5f4e77e5a8535f</MD5> <SHA256>e8597d7d5910a709985bffae490dfa9490483eb7838b351191156f027d9b5272</SHA256> <FileName>pts-sample-photos-2.tar.bz2</FileName> <FileSize>29599729</FileSize> </Package> </Downloads> </PhoronixTestSuite>
install.sh
#!/bin/bash chmod +x mistral-7b-instruct-v0.2.Q8_0.llamafile chmod +x llava-v1.5-7b-q4.llamafile chmod +x wizardcoder-python-34b-v1.0.Q6_K.llamafile tar -xf pts-sample-photos-2.tar.bz2 echo $? > ~/install-exit-status cat <<'EOT' > run-mistral #!/bin/bash ./mistral-7b-instruct-v0.2.Q8_0.llamafile --temp 0.7 -p '[INST]Write a long story about llamas[/INST]' $@ exit $? EOT chmod +x run-mistral cat <<'EOT' > run-wizardcoder #!/bin/bash ./wizardcoder-python-34b-v1.0.Q6_K.llamafile --temp 0 -e -r '```\n' -p '```c\nvoid *memcpy_sse2(char *dst, const char *src, size_t size) {\n' $@ exit $? EOT chmod +x run-wizardcoder cat <<'EOT' > run-llava #!/bin/bash ./llava-v1.5-7b-q4.llamafile --temp 0.2 --image DSC_4646.JPG -e -p '### User: Describe in detail what do you see?\n### Assistant:' $@ exit $? EOT chmod +x run-llava echo "#!/bin/sh ./\$@ -t \$NUM_CPU_PHYSICAL_CORES > \$LOG_FILE 2>&1 echo \$? > ~/test-exit-status" > ~/llamafile chmod +x ~/llamafile
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>Llamafile</Title> <AppVersion>0.6</AppVersion> <Description>Mozilla's Llamafile allows distributing and running large language models (LLMs) as a single file. Llamafile aims to make open-source LLMs more accessible to developers and users. Llamafile supports a variety of models, CPUs and GPUs, and other options.</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> <EnvironmentSize>40000</EnvironmentSize> <ProjectURL>https://llamafile.ai/</ProjectURL> <RepositoryURL>https://github.com/Mozilla-Ocho/llamafile</RepositoryURL> <Maintainer>Michael Larabel</Maintainer> </TestProfile> <TestSettings> <Option> <DisplayName>Test</DisplayName> <Identifier>test</Identifier> <Menu> <Entry> <Name>mistral-7b-instruct-v0.2.Q8_0</Name> <Value>run-mistral</Value> </Entry> <Entry> <Name>llava-v1.5-7b-q4</Name> <Value>run-llava</Value> </Entry> <Entry> <Name>wizardcoder-python-34b-v1.0.Q6_K</Name> <Value>run-wizardcoder</Value> </Entry> </Menu> </Option> <Option> <DisplayName>Acceleration</DisplayName> <Identifier>accel</Identifier> <Menu> <Entry> <Name>CPU</Name> <Value>--gpu DISABLE</Value> </Entry> <Entry> <Name>GPU AUTO</Name> <Value>--gpu AUTO</Value> </Entry> </Menu> </Option> </TestSettings> </PhoronixTestSuite>