If you want to run the Docker image in a complete offline environment, you need to add the `--build-arg with_models=true` parameter. Then the language models are downloaded during the build process of the image. Otherwise, these models get downloaded on the first run of the image/container.
Run the built image:
```bash
docker run -it -p 5000:5000 libretranslate [args]
```
Or build and run using Docker Compose:
```bash
docker compose up -d --build
```
> Feel free to change the [`docker-compose.yml`](https://github.com/LibreTranslate/LibreTranslate/blob/main/docker-compose.yml) file to adapt it to your deployment needs, or use an extra `docker-compose.prod.yml` file for your deployment configuration.
>
> The models are stored inside the container under `/home/libretranslate/.local/share` and `/home/libretranslate/.local/cache`. Feel free to use volumes if you do not want to redownload the models when the container is destroyed. To update the models, use the `--update-models` argument.
Some users may encounter the following error while installing packages:
```
error: externally-managed-environment
× This environment is externally managed
╰─> To install Python packages system-wide, try apt install
python3-xyz, where xyz is the package you are trying to
install.
…
```
This occurs when your operating system depends on and manages Python for core functionality. In this case, you should install and setup venv(virtual environments) to manage project dependencies.
This prevents pip packages from being installed system-wide. This way, there are no risks of pip packages conflicting between multiple projects or the operating system.
