Polish LLM benchmark is a compact yet powerful platform designed for benchmarking in langue open-source large language models (LLMs) on polish language. Equipped with diverse test datasets, it facilitates rigorous and comprehensive evaluation of LLMs. This tool streamlines the process of assessing model performance, enabling researchers and developers to reliably test and share their findings in the dynamic landscape of AI and natural language processing.
This section will guide you through getting a copy of the project up and running on your local machine for development and testing purposes. Before you start, ensure you have Git installed on your system. If not, you can download and install it from Git's official website.
- Git installed on your machine.
- Docker installed for running the application in a container (optional but recommended).
Clone the repository to get a local copy on your machine, Change into the project directory, Run the setup script to configure the environment.
git clone https://github.com/knmlprz/Benchmark_LLM.git
cd Benchmark_LLM
./set_venv.sh
source venv/bin/activate
Clone the repository to get a local copy on your machine, Change into the project directory, Run the setup script to configure the environment
git clone https://github.com/knmlprz/Benchmark_LLM.git
cd Benchmark_LLM
set_venv.bat
venv\Scripts\activate.bat
First Time Setup
Build and start the Docker containers. This step is required only once or after any changes in the Docker configuration.
docker compose up --build
Regular Usage
Start the Docker containers without rebuilding.
docker compose up
Shutdown
To stop and remove the containers, networks, and volumes, use:
docker compose down
First Time Setup
Build and start the Docker containers. This step is required only once or after any changes in the Docker configuration.
docker compose up --build
Regular Usage
Start the Docker containers without rebuilding.
docker compose up
Shutdown
To stop and remove the containers, networks, and volumes, use:
docker compose down