What it does

AnPyer runs notebook cells with an IPython kernel inside the app. The interpreter is CPython 3.13 running on the device. It is not a server in a browser tab or a remote sandbox. You open any .ipynb file, run cells, and get tables and plots. NumPy, pandas, SciPy and 70+ packages are preinstalled. The package list is fixed per release, and there is no pip install. You can grant access to the folder that holds your notebook. Files that Python writes are synced back to that folder. An optional AI assistant works inside the notebook and uses an API key from an OpenAI-compatible provider. AnPyer requires Android 8.0+ and a 64-bit ARM device.

Capabilities

The IPython kernel supports tab completion, ? help, magics, interrupt and restart.
The output renderer shows Markdown, LaTeX math, Mermaid diagrams, Graphviz DOT, chemical structures (SMILES), interactive ECharts and plotly. It renders them in Markdown cells and in cell output.
The AI assistant reads cells and their outputs, and it edits or inserts cells. Every edit shows a confirmation card, and every accepted change is one Undo step.
The AI assistant runs a cell only if you enable it for the session. The permission is off by default and off again when you relaunch the app.
You can mark a cell as hidden or locked. AnPyer says a hidden cell gives the assistant no source, no outputs and no hash. AnPyer says the assistant can read a locked cell but never change it.

When to use it

You want to run Python notebook cells on an Android phone or tablet, including offline.
You want to process your own data from a folder on the device.
You write a notebook on a laptop, edit it on the phone, and open it again on the laptop.