# Omnichar

> Documentation for Omnichar: build a character once as a portable .char file and use it in Omnichar Cloud, the open-source Omnichar Studio, or ComfyUI.

- [About Omnichar](https://docs.omnichar.org/index.md): Omnichar keeps the same face, body, clothes and voice across shots and models. Build a character once and use it in the cloud, locally or in ComfyUI.
- [The .char character file](https://docs.omnichar.org/characters.md): What a .char file holds, which models take a character, and how one file moves between Omnichar Cloud, Omnichar Studio and ComfyUI.
- [Omnichar Cloud overview](https://docs.omnichar.org/cloud/overview.md): Omnichar Cloud runs in your browser: sign in with Google, build a character from 1-3 photos, and generate images and video up to 4K with no GPU.
- [Omnichar Cloud pricing and credits](https://docs.omnichar.org/cloud/pricing.md): What Omnichar Cloud charges: one credit is one cent, images and video have a fixed price shown before you run, and failed runs are never charged.
- [Omnichar Studio overview](https://docs.omnichar.org/app/overview.md): Omnichar Studio is the free, open-source app for AI characters: a node canvas with take history, LoRA training and a local engine, running on your own GPU.
- [Install Omnichar Studio](https://docs.omnichar.org/app/install.md): Install Omnichar Studio on Linux, macOS or Windows with Python 3.11+. One command sets up the engine, the model runtime, the LoRA trainer and the web UI.
- [Supported models in Omnichar Studio](https://docs.omnichar.org/app/models.md): The local models Omnichar Studio runs and trains, from Z-Image Turbo to MiniMax H3 and LTX-2.5, with VRAM needs and where to put the model files.
- [Omnichar Studio command-line options](https://docs.omnichar.org/app/command-line.md): Flags and INLINE_* environment variables for webui.sh and webui.bat: port, LAN access, model folder, low-VRAM mode, multi-GPU and PyTorch index.
- [Install the Omnichar ComfyUI node](https://docs.omnichar.org/comfyui/install.md): Install ComfyUI-Omnichar to load .char characters into your ComfyUI workflows, with consistent faces, bodies, clothes and voice on FLUX.2, Krea 2 and MiniMax H3.
- [Omnichar ComfyUI nodes reference](https://docs.omnichar.org/comfyui/nodes.md): Inputs and outputs of every Omnichar ComfyUI node: Load, Decode, Encode and Save Character, the reference nodes, and Apply Character LoRA.
- [Omnichar ComfyUI workflows](https://docs.omnichar.org/comfyui/workflows.md): Example ComfyUI workflows for Omnichar: build a .char from reference photos, render it with FLUX.2 Klein or MiniMax H3, and give a character a voice.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.