OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Image & Video AI

Omnichar

OmniChar is an open-source AI character studio that stores face, body, wardrobe and voice information in portable .char files for reuse across generation shots and models.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Omnichar — OpenViST project profile in Image & Video AI.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use OmniChar when a character should remain recognizable across multiple AI-generated shots or models, and a portable profile is useful to the workflow.

Who it is for

  • Creators who need recurring characters to retain appearance across generated images or video shots.
  • ComfyUI users looking to bring character profiles into creative workflows.
  • Users choosing between local GPU generation and hosted-model generation.

Documented uses

  • Create a reusable character profile containing appearance and voice traits.
  • Maintain a consistent cast member across multiple shots and models.
  • Use character information in ComfyUI-based workflows.
  • Generate with hosted models when a local GPU is not being used.
Before you choose

Requirements & limits

The repository describes a browser-based option and a local Python 3.11+ installation. Local setup instructions are provided for macOS/Linux and Windows; hosted models are listed as a generation route that does not require a local GPU.

  • The .char format groups face, body, wardrobe and voice information in one portable profile.
  • Supported models differ in whether they support training, generation or both; hosted-model API nodes are listed for generation only.
  • Local model memory requirements vary. The README cautions that not every 16 GB training result was run on a 16 GB card.
  • A ComfyUI node is referenced, but specific node behavior is not described in the available repository text.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up the local web UI on macOS or Linux

Install OmniChar's local web UI and start it at the documented local address.

Before you start

  • Python 3.11 or later
  • A macOS or Linux environment with Git and a shell

Steps

  1. Clone the OmniChar repository and change into its core directory.

  2. Run ./webui.sh --install --extra all to install the engine, model runtime, trainer and UI.

  3. Run ./webui.sh to start the web UI.

Expected outcome

The README specifies the local web UI address as http://127.0.0.1:8848.

Guide limitations

  • This procedure documents setup and UI launch, not a particular character-generation task.
  • Local generation depends on model and hardware requirements; the README lists requirements separately by model.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
14 October 2026
Original website title
omnichar/OmniChar · Official homepage title · retained in its source language
How we review sourcesSuggest a correction