InferencePort AI Help

Welcome to the beginner docs for InferencePort AI.

InferencePort AI is a desktop app for local AI usage. You can chat with models, download or remove models, save sessions, browse Spaces/websites, and optionally sign in for account and sync features.

Start Here (5 Minutes)

1. Use Download Guide to pick the right installer for your OS.

2. Follow Getting Started for install and first launch.

3. Open User Guide for full feature-by-feature usage.

Quick Help

If you need one answer quickly, start here:

• I want my first working chat: Getting Started

• I want to download or remove models: User Guide

• I want to import .gguf or Modelfile models: User Guide

• I want to use Hugging Face Spaces or websites: User Guide

• I want optional sign-in and sync: User Guide

• I want remote host or hosting/proxy setup: User Guide

• Something is broken: Troubleshooting

What You Can Do In The App

• Local chat with streaming responses

• Model management (download, run, remove)

• Session management (continue, rename, delete, favorites)

• Lightning mode support

• GGUF/Modelfile imports

• Hugging Face Spaces and website cards

• Optional account features (email, Google, GitHub)

• Optional chat sync

• Remote host and local hosting/proxy controls

• Logs and diagnostics

Complete Documentation

End-User Docs

• Overview
• Getting Started
• Download Guide
• User Guide
• Troubleshooting
• Integrations

Developer Docs

• Developer Guide
  • Lightning API
  • Tools Workflow
  • GGUF Import Guide
  • Hugging Face Import Guide
  • Security and Privacy

Contributing Docs

• Contributing Docs
  • Contributing
  • Architecture
  • IPC API Reference
  • Development Guide
  • Build and Release
  • Security and Privacy (Internals)

Indices and tables

• Index

• Search Page