Documentation
What these pages cover.
Short pages for someone who wants to run KonfAI: what to install, what a config says, what each command writes, how data is laid out, and how the apps, the MCP server, Scout, Studio and Edge fit around the engine. Every page ends where the full reference begins; nothing here replaces it.
Get started
Install KonfAI and run the Segmentation example.
Configuration
One YAML per workflow: roots, dataset, groups, transforms, patches, models.
Workflows
TRAIN, RESUME, PREDICTION, EVALUATION, TRANSFORM and the options that matter.
Datasets
The layout KonfAI reads and the formats it reads regionally.
Apps
Published bundles, Hugging Face, the one-line CLI, 3D Slicer.
Agents and MCP
What the MCP server lets an agent do.
Scout
A visual inspector of results.
Studio
A chat web UI over the MCP server.
Edge
KonfAI apps in a browser tab, no server.
Which page first
- You have images and want a model: Get started, then Configuration.
- You want a result without training: Apps.
- You want to prepare data only: TRANSFORM and Datasets.
- You drive KonfAI from an agent or from a chat: Agents and MCP, Studio.
- You want to look at a result the way a radiologist would: Scout.
- You want it in a browser tab, no server: Edge.