This repo contains the Parallel Code Evaluation (ParEval) Benchmark for evaluating the ability of Large Language Models to write parallel code. See the ParEval Leaderboard for up-to-date results on different LLMs.
The organization of the repo is as follows.
prompts/
-- the prompts in ParEval alongside some utility scriptsgenerate/
-- scripts for generating LLM outputsdrivers/
-- scripts to evaluate LLM outputsanalysis/
-- scripts to analyze driver results and compute metricstpl/
-- git submodule dependencies
Each subdirectory has further documentation on its contents. The general
workflow is to use generate/generate.py
to generate LLM outputs, run
drivers/run-all.py
to evaluate outputs, and analysis/metrics.py
to
post-process the results.
A couple core systems software are assumed to be installed: Python >=3.7, a C++ compiler that supports C++17 and OpenMP, Make, CMake, and an MPI implementation. If you are testing the CUDA and HIP prompts, then you will need access to NVIDIA and AMD GPUs alongside their respective software stacks.
First, clone the repo.
git clone --recurse-submodules https://github.com/parallelcodefoundry/ParEval.git
Next, you need to build Kokkos (if you want to include it in testing).
cd tpl/kokkos
mkdir build
cd build
# depending on your system you may need to pass your c++ compiler to CMAKE_CXX_COMPILER
cmake .. -DCMAKE_INSTALL_PREFIX=. -DKokkos_ENABLE_THREADS=ON
make install -j4
Finally, you need to install the Python dependencies. requirements.txt
has
the set of dependencies pinned at the version they were tested with. Other
versions may also work. Note that some of these are only required for parts of
the pipeline i.e. PyTorch and Transformers are only needed for generating LLM
outputs.
pip install -r requirements.txt
@misc{nichols2024large,
title={Can Large Language Models Write Parallel Code?},
author={Daniel Nichols and Joshua H. Davis and Zhaojun Xie and
Arjun Rajaram and Abhinav Bhatele},
year={2024},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
booktitle = {Proceedings of the 33rd International Symposium on High-Performance Parallel and Distributed Computing},
series = {HPDC '24}
}
ParEval is distributed under the terms of the MIT license.