---

8.1 Installing a Large Language Model ¶

We don’t recommend to use machine translation through a web service in the cloud, controlled by someone else than yourself. Such a machine translation service would be have major drawbacks (it could go away any time, it could be used to spy on you or manipulate you, or the costs could go up beyond your control); see https://www.gnu.org/philosophy/who-does-that-server-really-serve.en.html. Additionally, such a service typically has some cost (between 10ドル and 25ドル per megabyte, as of 2025).

Instead, we recommend a Large Language Model execution engine that runs on hardware under your control. This can be a desktop computer, or for instance a single-board computer in your local network.

At this point (in 2025), a Large Language Model execution engine that is Free Software is ‘ollama’, that can be downloaded from https://ollama.com/.

Next, you will need to pick a Large Language Model. There are two properties to watch out for:

• The license under which the LLM is released. While some LLMs are under a Free Software license (most often the Apache 2.0 license) and are thus “open weights” models (see https://opensource.org/ai/open-weights), others have licenses with usage restrictions.
• The size of the LLM. Generally, it should be smaller than the available RAM size of your machine on which you will run the LLM execution engine.

Together with an LLM of reasonable quality, such as the model ministral-3:14b, the system requirements are as follows:

• RAM: 16 GB.
• Disk space: 10 GB (1 GB for ollama, 9 GB for the model).
• GPU or TPU: A GPU (Graphics Processing Unit) or TPU (Tensor Processing Unit) is not needed. As of 2025, a high-end GPU from certain vendors can be used by ollama to provide an optional speedup.

Additional configuration:

• If you are running ollama on your computer directly, no further configuration is needed.
• If you are running ollama on a separate machine, and want to make it accessible from all machines in the LAN: Edit the file /etc/systemd/system/ollama.services, adding a line: Environment="OLLAMA_HOST=0.0.0.0". See https://github.com/ollama/ollama/issues/703.
• If you are running ollama in a virtual machine, make the port 11434 accessible through port forwarding.

Next: Invoking the msgpre Program, Up: Pretranslating PO Files [Contents][Index]