OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Infrastructure & Serving

Deepstack

DeepStack is an AI API engine that serves pre-built and custom computer-vision models locally on edge devices or in a private cloud.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Deepstack — OpenViST project profile in AI Infrastructure & Serving.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use DeepStack when an application needs a locally hosted computer-vision API, including on edge hardware or a private cloud VM.

Who it is for

  • Developers building applications that need locally served face, object or scene analysis.
  • Teams deploying computer-vision services on edge devices or private infrastructure.

Documented uses

  • Detect faces, recognize or match faces, and detect common objects in images.
  • Deploy custom models for detecting selected object classes.
  • Classify image scenes or enhance images with 4× super-resolution.
  • Serve computer-vision models from an edge device or private cloud.
Before you choose

Requirements & limits

  • The README lists Linux and macOS through Docker, Windows 10 as a native application, and NVIDIA Jetson and Raspberry Pi or ARM64 devices through Docker.
  • The project describes offline operation independent of the cloud.
  • The repository provides CPU, GPU and Jetson Docker build targets.
  • Building the Docker version from source requires Go, Docker, Git, Git LFS, Redis, Python 3.7 and PowerShell 7 or later.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Build a DeepStack Docker image from source

Prepare the repository and build a CPU, GPU, or Jetson Docker image.

Before you start

  • Golang
  • Docker
  • Git and Git LFS
  • Redis Server
  • Python 3.7

Steps

  1. Install the listed prerequisites.

  2. Clone the repository with the command below, then enter it with cd DeepStack.

    Documented command
    git clone https://github.com/johnolafenwa/DeepStack.git

  3. Fetch repository files with the command below and download binary dependencies using .\download_dependencies.ps1.

    Documented command
    git lfs pull

  4. Build one image from the repository’s parent directory: CPU with sudo docker build -t deepquestai/deepstack:cpu . -f Dockerfile.cpu, GPU with sudo docker build -t deepquestai/deepstack:gpu . -f Dockerfile.gpu, or Jetson with sudo docker build -t deepquestai/deepstack:jetpack . -f Dockerfile.gpu-jetpack.

Expected outcome

A Docker image is built for the selected CPU, GPU, or Jetson target.

Guide limitations

  • This recipe documents image building, not running or testing the resulting image.
  • The listed steps specify Docker-version builds; they do not document installation instructions for every supported platform.
Documentation 1 ↗
Source-based practical guide

Run DeepStack locally without building a Docker image

Set up and start the local server with CPU or GPU dependencies and selected vision APIs enabled.

Before you start

  • Python 3.7
  • PowerShell 7 or later
  • The repository’s listed prerequisites: Go, Docker, Git, Git LFS, and Redis Server

Steps

  1. Create and activate a virtual environment with the command below and source venv/bin/activate, then install requirements using the command below.

    Documented command
    python3.7 -m venv venv
    Documented command
    pip3 install -r requirements.txt

  2. Install the documented PyTorch packages for your target: CPU with the command below, or GPU with the command below.

    Documented command
    pip3 install torch==1.6.0+cpu torchvision==0.7.0+cpu -f https://download.pytorch.org/whl/torch_stable.html
    Documented command
    pip3 install torch==1.6.0+cu101 torchvision==0.7.0+cu101 -f https://download.pytorch.org/whl/torch_stable.html

  3. Start PowerShell, then run .\setup_docker_cpu.ps1 for CPU or .\setup_docker_gpu.ps1 for GPU.

  4. Change to the server directory and build the server with the command below. Set one or more documented API flags to True: VISION_DETECTION, VISION_FACE, or VISION_SCENE; then start the server with .\server.

    Documented command
    go build

Expected outcome

The README says DeepStack runs locally on default port 5000 and that logs are in the repository-root `directory`.

Guide limitations

  • The README documents CPU and GPU local setup, not a complete verification procedure or client/API request example.
  • The commands are presented as written in the README; it does not explain how to configure other server options.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
15 October 2026
Original website title
johnolafenwa/DeepStack · Official homepage title · retained in its source language
How we review sourcesSuggest a correction