An IPython extension for discussing your interactive Python session with models
supported by the llm Python package. For a detailed description,
as well as a discussion of use cases, please refer to the blog.
Install ipyllm from PyPI into the Python environment used by IPython:
python -m pip install ipyllmOr install from a local checkout:
python -m pip install .Then load it in IPython or a Jupyter IPython kernel:
%load_ext ipyllm
The extension uses the model and API-key configuration already managed by
llm. For example, choose a default model with llm models default MODEL and
configure provider keys using llm keys set PROVIDER.
The MkDocs documentation covers installation, model
configuration, prompts, adding files to the context, command-output and
last-error diagnosis, debugging with ipyllm.embed, available tools, Agent
Skills, per-prompt and per-model token usage,
approval settings, and safety. Preview or build the site with:
uv run mkdocs serve
uv run mkdocs buildAn online version is available at the homepage.