Skip to content

Quickstart

This guide takes a beginner from PyPI install to first validation.


Requirements

Tool Required
Python 3.11+ recommended, 3.12 preferred
Git Yes
Docker Optional but recommended
DeepSeek/OpenAI key Optional for live provider tests

1. Install from PyPI

python -m pip install -U pip
pip install largestack
largestack --help
  • PyPI: https://pypi.org/project/largestack/
  • GitHub: https://github.com/Rivailabs/largestack

2. Source development checkout

Use this only when you want to work on the framework source:

git clone https://github.com/Rivailabs/largestack.git
cd largestack

3. Create virtual environment

Linux/macOS:

python3.12 -m venv .venv
source .venv/bin/activate

Windows PowerShell:

py -3.12 -m venv .venv-win
.\.venv-win\Scripts\Activate.ps1

4. Install development dependencies

python -m pip install -U pip setuptools wheel
python -m pip install -e ".[dev]"

If PyTorch tries to install large CUDA packages on Linux, use CPU wheels:

PIP_EXTRA_INDEX_URL=https://download.pytorch.org/whl/cpu \
python -m pip install -e ".[dev]"

5. First test

python -m pytest tests/unit/test_memory.py -q --tb=short

Expected:

10 passed

6. Compile check

python -m compileall largestack tests examples scripts

This catches syntax/import-path problems early.


7. Full test suite

python -m pytest tests -q --tb=short -ra

Live DeepSeek tests run only when LARGESTACK_DEEPSEEK_API_KEY is set.


8. Live DeepSeek check

export LARGESTACK_DEEPSEEK_API_KEY="your_key_here"
python -m pytest tests/integration/test_deepseek_integration.py -q --tb=short -ra

Never commit keys.


9. Docker health check

docker build -t largestack:test .
docker run --rm -d --name largestack-test -p 8787:8787 largestack:test
curl http://127.0.0.1:8787/health
docker rm -f largestack-test

Expected health response should show an OK/healthy status.

Status

Largestack AI is suitable for developer demos, controlled pilots, AI workflow experiments, and early enterprise evaluation. It is not currently presented as a certified enterprise SaaS platform.