,______  .______ .______  ,___
: __   \ \____  |:      \ : __|
|  \____|/  ____||  _,_  || : |
|   :  \ \   .  ||   :   ||   |
|   |___\ \__:__||___|   ||   |
|___|        :       |___||___|
             *

Run r2ai in local, without internet or leaking any data. Use your CPU/GPU/NPU and interact with r2 using natural language. The current implementation is based on llama and the default model is CodeLlama-CherryPop

--pancake

• Prompt the language model without internet requirements
• Slurp file contents and make actions on that
• Embed the output of an r2 command and ask the LLM to resolve questions
• Define different system-level assistant role
• Set environment variables to provide context to the language model
• Live with repl and batch mode from cli or r2 prompt
• Accessible as an r2lang-python plugin, keeps session state inside radare2
• Scriptable from bash, r2pipe, and javascript (r2papi)
• Use different models, dynamically adjust query template
  • Load multiple models and make them talk between them

The easiest way to run and install r2ai is by installing the latest r2 from git and run these lines:

r2pm -i rlang-python # optional
pip3 install rich inquirer llama-cpp tokentrim hugging_face appdirs
r2pm -ci r2ai

On recent Debian/Ubuntu systems the pip tool is no longer working, because it conflicts with the system packages. The best way to do this is with venv:

python -m venv r2ai
. r2ai/bin/activate
pip install rich inquirer llama-cpp-python tokentrim hugging_face appdirs
r2pm -r r2ai

On native Windows follow these instructions (no need to install radare2 or use r2pm), note that you need Python 3.8 or higher:

git clone https://github.com/radareorg/r2ai
cd r2ai
set PATH=C:\Users\YOURUSERNAME\Local\Programs\Python\Python39\;%PATH%
python -m pip install rich inquirer llama-cpp-python tokentrim hugging_face appdirs pyreadline3
python main.py

There are 4 different ways to run r2ai:

• Standalone and interactive: r2pm -r r2ai
• Batch mode: r2ai '-r act as a calculator' '3+3=?'
• From radare2 (requires r2pm -ci rlang-python): r2 -c 'r2ai -h'
• Using r2pipe: #!pipe python main.py

You can interact with r2ai from standalone python, from r2pipe via r2 keeping a global state or using the javascript intrepreter embedded inside radare2.

• conversation.r2.js - load two models and make them talk to each other

Just run make .. or well python main.py /path/to/file

It's also possible to install it with conda, which is the recommended way on Macs.

• https://developer.apple.com/metal/pytorch/

curl -O https://repo.anaconda.com/miniconda/Miniconda3-latest-MacOSX-arm64.sh
sh Miniconda3-latest-MacOSX-arm64.sh

conda install pytorch torchvision torchaudio -c pytorch-nightly
conda run pip install inquirer rich appdirs huggingface_hub tokentrim llama-cpp-python

• add "undo" command to drop the last message
• dump / restore conversational states
• custom prompt templates