README.md

---
title: Inline Studio
---

# Inline Studio

Free, open-source AI filmmaking on a node canvas. Build a whole visual pipeline, from moodboard to final cut, with local diffusion models and hosted models, and train your own LoRAs on the same canvas. Every render is kept as a versioned, non-destructive take.

This Hugging Face org holds models and datasets trained with the app.

- Website: https://inlinestudio.art
- App and source: https://github.com/inlineresearch/Inline-Studio
- Getting started: https://inlinestudio.art/getting-started

## What you will find here

- **LoRAs** trained in the Inline Studio Trainer tab, ready to drop into the app (or any Krea 2 / Z-Image workflow).
- **Datasets** used to train them: image and caption pairs you can reuse or learn from.

Current releases:

- [skin-lora-krea-2-raw](https://huggingface.co/inlineresearch/skin-lora-krea-2-raw) - a photorealistic skin-texture LoRA for Krea 2.
- [krea2-skin-lora](https://huggingface.co/datasets/inlineresearch/krea2-skin-lora) - the dataset it trained on.

## About the app

Inline Studio runs as one process on one port: the built-in Inline Core engine serves the web UI and does the generation. Krea 2 and Z-Image run locally on your own GPU from a single model file, LoRA training is a second canvas of nodes (dataset, captioning, training, loss curve), and the finished LoRA drops straight back into the app to generate with.

GPL-3.0. Contributions welcome on [GitHub](https://github.com/inlineresearch/Inline-Studio).