Skip to main content
Version: 0.10.x [Latest Beta]

Integration overview

This chapter covers installing, configuring, and operating the DENKflow SDK from Python and C/C++.

Fastest path to first inference​

  1. Export a .denkflow file from the Vision AI Hub.
  2. Request and activate a license.
  3. Install the SDK for the target runtime and language.
  4. Run the export through SimplifiedPipeline and read the results.

The Quick start contains the complete first-inference example.

How this chapter is organized​

  • Getting started: installation, licensing, models, and a first inference
  • Using SimplifiedPipeline: the default inference API, its inputs, results, and result drawing
  • Examples: task-oriented examples per model type
  • Deployment: runtime selection, configuration, containers, and troubleshooting
  • Advanced usage: the classic Pipeline flow, custom graph construction, and tensors
  • Reference: changelog, migration guides, and the LLM skills file

Choose a path​

Running exported pipelines​

Use a .denkflow export with:

  1. Quick start
  2. Installation guide
  3. Authentication and licensing
  4. SimplifiedPipeline

Building pipelines from .denkmodel files​

Assemble graphs manually with:

  1. Installation guide
  2. Pipeline fundamentals
  3. Creating ImageTensors
  4. Custom pipeline construction

Deploying in containers​

  1. Configuration
  2. Docker deployment
  3. Troubleshooting

Upgrading​