Skip to main content
Version: 0.5.x [Latest Alpha]

Installation Guide

This guide will help you set up Denkflow on your system. Denkflow is optimized for AI acceleration on compatible hardware but will work on standard CPU systems as well.

System Requirements​

Minimum Requirements​

  • Operating System: Windows 10, Windows 11, Linux (with glibc ≥ 2.31 and libstdc++ ≥ 12)
  • Architecture: x86-64 or ARM64
  • Storage: At least 2GB of free disk space for models and cache
  • RAM: 4GB minimum, 16GB+ recommended for larger models
  • Optional GPU: NVIDIA GPU or Jetson Device with 4GB+ VRAM for acceleration
  • CPU: 4+ cores for optimal performance

Standard Installation​

Prerequisites​

  • Python 3.10 or newer

Installation​

For a basic installation that will run on CPU:

pip install denkflow --index-url https://alice:glpat-8ZB7unxdFxiTGdW-BBzA@gitlab.com/api/v4/projects/69262737/packages/pypi/simple

Hardware Acceleration Options​

Denkflow supports various hardware acceleration methods to significantly improve inference speed. Choose the option that matches your hardware:

CUDA (NVIDIA GPUs)​

For accelerated processing on NVIDIA graphics cards:

pip install denkflow[cuda] --index-url https://alice:glpat-8ZB7unxdFxiTGdW-BBzA@gitlab.com/api/v4/projects/69262737/packages/pypi/simple

Requirements:

  • CUDA Toolkit version 12.x installed on your system
  • Compatible NVIDIA GPU (GTX 1060 6GB or better recommended)
  • Up-to-date NVIDIA drivers

TensorRT (NVIDIA Jetson & GPUs)​

For maximum inference speed on NVIDIA Jetson devices and compatible GPUs:

pip install denkflow[tensorrt] --index-url https://alice:glpat-8ZB7unxdFxiTGdW-BBzA@gitlab.com/api/v4/projects/69262737/packages/pypi/simple

Notes:

  • Requires NVIDIA TensorRT version 10.x installed on your system
  • First-time initialization takes ~15 minutes to build the TensorRT engine cache
  • Subsequent runs will be significantly faster
  • The cache is stored in the DENKFLOW_DATA_DIRECTORY (see Configuration section)

Troubleshooting​

If you encounter installation issues:

  1. Ensure your Python version meets requirements: python --version
  2. Check your system libraries: ldd --version
  3. For GPU acceleration, verify CUDA installation: nvcc --version
  4. See our Troubleshooting Guide for common solutions

Next Steps​

After successful installation, check out the Core Concepts to begin using Denkflow.