KonfAI KonfAI KonfAI
/

Getting started

  • Installation
  • Quickstart

Tutorials

  • Examples
    • KonfAI on real data
    • Segmentation Example
    • Registration example
    • Synthesis Example

Task-oriented guides

  • How-to guides
    • Adopt KonfAI from PyTorch, MONAI, or nnU-Net
    • Processing large images
    • Using custom models, transforms, augmentations, and losses
    • Docker

Core concepts

  • Core concepts
    • Configuration model
    • Datasets and groups
    • Patch streaming
    • Model graph and output naming
    • Declarative YAML model graphs
    • Execution flow
  • Configuration reference
    • Training configuration
    • Prediction configuration
    • Evaluation configuration

Apps and automation

  • Using KonfAI Apps
  • Agent workflows (MCP server)
  • Ecosystem

Reference

  • Reference
    • Built-in component catalogue
      • Models
      • Losses & metrics
      • Transforms
      • Augmentations
      • Schedulers
      • Storage backends & formats
    • CLI reference
    • Environment variables
    • App server HTTP API
    • Python API (apps)
    • API reference
      • Workflows API
      • Configuration API
      • Data API
      • Models API
      • Apps API
      • Extension points
      • Full module reference
        • konfai package

Project

  • Troubleshooting
  • Development and contributing
  1. KonfAI /
  2. Reference /
  3. API reference

API reference#

This section is split into two levels:

  • Curated API pages for the public building blocks most developers interact with

  • a full module reference generated from the package modules

The pages below combine short explanations with API signatures pulled directly from the code docstrings.

  • Workflows API
  • Configuration API
  • Data API
  • Models API
  • Apps API
  • Extension points
  • Full module reference
Previous
Python API (apps)
Next
Workflows API

2026, Valentin Boussot

Made with Sphinx and Shibuya theme.