ipy-llm-kernel is a Jupyter kernel that allows you to type in English language and receive responses from a large language model (LLM).

It can make use of OpenAI's chatGPT, Anthropic's Claude, Helmholtz' blablador and Ollama. You need an OpenAI API, Anthropic, Google or a Helmholtz account to use it. Using it with Ollama is free but requires running an Ollama server locally.

After starting jupyter lab, select the LLM Kernel.

You can then type in English language and receive responses from the LLM as demonstrated above

If your prompt asks for generating images, it may generate this image for you. At the moment, only OpenAI/Dall-E3 is supported for this. You need an OpenAI API Key for this.

First, you should also create an environment variable named "IPY_LLM_KERNEL_MODEL" and enter a model name depending on which service provider you want to use. Examples:

• llama3:8b
• blablador:alias:large
• claude-3-5-sonnet-20240620
• gpt-4o-2024-08-06

Then, start a new terminal to install ipy-llm-kernel using pip. It is recommended to install it into via conda/mamba environment. If you have never used conda before, please read this guide first.

pip install ipy-llm-kernel

Afterwards, run additionally this command:

python -m ipy_llm_kernel install

You can check if it's installed by printing out the list of installed kernels:

jupyter kernelspec list

And you can uninstall them using this command:

jupyter kernelspec uninstall llm-kernel

If you want to contribute to ipy-llm-kernel, you can install it in development mode like this:

git clone https://github.com/haesleinhuepf/ipy-llm-kernel.git
cd ipy-llm-kernel
pip install -e .

There are similar projects:

• jupyter-ai
• chatGPT-jupyter-extension
• chapyter
• ipython-gpt

If you encounter any problems or want to provide feedback or suggestions, please create an issue along with a detailed description and tag @haesleinhuepf .