Lhammai CLI allows you to interact with any LLM directly from your terminal using a simple, intuitive interface.
Powered by the any-llm library, it seamlessly connects to various LLM
providers, including OpenAI, Anthropic, and local servers such as Ollama, llamafile, and others. For a full list of
supported providers, see the official any-llm documentation.
The name Lhammai comes from "Lhammas," Noldorin for "account of tongues", a work of fictional sociolinguistics, written by J. R. R. Tolkien in 1937.
You can install the package from PyPI using pip (recommended):
pip install "lhammai-cli[ollama]"-
Clone the repository and navigate to the source directory:
git clone https://github.com/dpoulopoulos/lhammai-cli.git && cd lhammai-cli -
Install the dependencies using
uv:uv sync --group ollama -
Activate the virtual environment:
source .venv/bin/activate
Note
This installs the necessary dependencies to communicate with a local model via Ollama.
To begin, you'll need to run the Ollama server. For this example, you can use Docker for a quick setup.
Warning
This approach has some limitations, especially on a Mac. Since Docker Desktop doesn't support GPUs, it's better to run Ollama as a standalone application if you're using a Mac. For more detailed instructions, check the official Ollama documentation.
-
Run the following command to start the Ollama server in a Docker container:
a. CPU only:
docker run -d -v ollama:/root/.ollama -p 11434:11434 --name ollama ollama/ollamab. Nvidia GPU:
docker run -d --gpus=all -v ollama:/root/.ollama -p 11434:11434 --name ollama ollama/ollama -
Run a model:
docker exec -it ollama ollama run gemma3:4b -
Interact with the model:
lhammai Hello!
Tip
Configure your application by creating a .env file in the root directory and adding your options:
cp .default.env .env
You can also pipe content to lhammai from standard input. This is useful for analyzing logs, summarizing files, etc.:
cat dev.log | lhammai -p "explain:"See the LICENSE file for details.