Developer Manual (Manual Install)

If you are a developer looking to contribute to Yumi, write custom tools, or tweak her LangGraph architecture, you should install her manually from source.

1. Prerequisites

Ensure you have the following tools installed:

Python 3.12+
Git
uv (Package Manager): The project uses Astral's uv instead of raw pip to manage dependencies.

If you don't have uv, install it:

# Windows (PowerShell)
powershell -c "irm https://astral.sh/uv/install.ps1 | iex"

# macOS & Linux
curl -LsSf https://astral.sh/uv/install.sh | sh

2. Clone & Sync

Clone the repository and sync the virtual environment. uv handles virtualenv creation automatically under the hood:

git clone https://github.com/CodeNeuron58/Yumi.git
cd Yumi
uv sync

CRITICAL RULE: Never use pip install! Yumi pins PyTorch and TorchAudio to a dedicated CPU-only index inside pyproject.toml (which reduces package size from 2 GB to ~220 MB). Using raw pip will bypass this config, attempting to download the full CUDA build, and will likely fail on Standard CPUs.

3. Activate the Environment

To make the global yumi command available in your current terminal session:

Windows (PowerShell)

.venv\Scripts\activate

macOS / Linux

source .venv/bin/activate

Once activated, you can execute yumi commands directly:

yumi --help
yumi wake-up

Alternatively, you can run commands without activation by prefixing with uv run:

uv run yumi

Proceed to WSL2 Installation or Core Senses to learn more!