VTuber Original Character Maker Launches, Turning One Image Into a Stream-Ready 3D Avatar

VTuber Original Character Maker officially launched on August 31, 2026 as a desktop application designed to shorten the process of creating a VTuber avatar. Users provide a single character illustration, and the software generates front, side and back views before automatically producing a rigged 3D model.

Developed by Sumeragi, the application extends beyond model generation. Expression sets, microphone-driven lip sync, context-sensitive expression switching, backgrounds, short videos and OBS output are all integrated into the same workflow.

Quick Summary

• Name: VTuber Original Character Maker
• Japanese name: VTuberオリキャラメーカー
• Type: VTuber avatar creation and streaming application
• Developer: Sumeragi
• Steam publisher: Sumeragi
• Official launch date: August 31, 2026
• Status: available through the official website
• Platforms: Windows and macOS
• Windows: Windows 10/11 64-bit
• macOS: macOS 14 Sonoma or later
• Steam: store page is live, but the software remains unreleased on Steam at the time checked
• Download price: free
• Business model: free core features with paid points for generation
• Core function: transforms one image into a 3D VTuber model
• Initial input: one character illustration or image
• Initial process: creates front, side and back views
• 3D model: generated automatically
• Rigging: handled automatically
• 3D character generation: 800 pt
• Expression sets: generated automatically
• Expression set cost: 200 pt
• Mouth shapes: six shapes covering closed, A, I, U, E and O
• Expression templates: include smile, angry and surprised
• Custom expressions: supported through natural-language descriptions
• Lip sync: real-time microphone-driven
• Expression switching: can respond to conversational context
• Speech processing: runs locally on the PC
• Microphone audio: not sent to a server for speech recognition or expression switching
• Background generation: supported
• Background cost: 60 pt
• Short video generation: supported
• Video cost: 85 pt per second
• Text-generated input image: available for 25 pt
• OBS: Browser Source output supported
• Additional OBS plugin: not required
• Bundled character: one ready-made 3D character is included
• Bundled-character streaming: can be tested without paid generation
• UI languages: Japanese, English and Simplified Chinese
• Internet connection: required for generation features
• Microphone: required for lip sync and voice features
• Official free-space requirement: at least 5 GB for the app, AI data and generated files
• First-launch AI data: approximately 1.5 GB
• Commercial use: generated characters may be used commercially or non-commercially under the terms
• Model export: generated 3D models can be exported
• Source-image rights: users must hold the necessary rights to their input image
• Steam AI disclosure: Steam lists a connection to Tripo for AI content generation

Trailer, Steam and Application Screenshots

One Illustration Becomes a Fully Rigged 3D Model

The main workflow begins with a single character image. The system analyzes that artwork and generates front, side and back views before using them as the foundation for the 3D model.

Rigging is handled automatically as well. The resulting character is therefore prepared for movement without requiring the user to manually build a skeleton in separate 3D software.

Expressions and Lip Sync Are Generated Too

Once the model exists, users can generate an expression set for 200 pt. Each set includes multiple expressions and six mouth shapes used for lip sync.

During streaming, microphone input drives the mouth in real time. Local AI can also analyze the conversation and switch the character's expression based on the subject or mood.

Sumeragi states that microphone audio and speech-recognition text used for these features remain on the PC. Around 1.5 GB of AI data is downloaded on first launch to enable the local speech and expression processing.

OBS Receives a Clean Feed Without the App UI

VTuber Original Character Maker provides dedicated OBS output. Enabling OBS Output creates a URL containing only the stream view, without the application's buttons or interface panels.

The URL can be pasted directly into OBS's standard Browser Source, with no additional plugin required. Window capture is also supported, while the microphone itself is added separately as an OBS audio source.

Free Download With Paid Generation Points

Downloading the software, using the bundled character, running lip sync, switching expressions and sending the view to OBS do not consume points.

Generating new assets is the paid portion. A 3D character costs 800 pt, an expression set 200 pt, a background 60 pt, video 85 pt per second, and a description-generated input image 25 pt.

A 600-point package costs ¥630, while the 2,000-point Standard Pack costs ¥2,080. Purchased points expire 180 days after they are granted.

Website Version Is Live While Steam Is Still Waiting

The Windows and macOS versions can be downloaded directly from the official website. Windows requires a 64-bit edition of Windows 10 or 11, while Mac users need macOS 14 Sonoma or newer.

The Steam page is also live with Sumeragi listed as both developer and publisher. At the latest check, however, Steam still marked the software as not yet available with its release date listed as To be announced.

Generated Models Can Be Used and Exported

The official FAQ states that generated characters and images can be used for both commercial and non-commercial purposes under the service terms. Users can also export a generated 3D model together with its animation and expression textures.

Users must, however, hold the necessary rights to the source artwork. The official site specifically warns against using another artist's illustration or a person's facial photograph without permission.

Game Ad

Launch Them All

Generative outputs are also not guaranteed to be exclusive, and the developer does not guarantee that an output will be free from potential third-party rights issues.

Why This Matters

1) It consolidates the VTuber creation workflow

Model generation, rigging, expressions, lip sync, backgrounds and OBS output are handled within one application.

2) Users do not need rigging experience to start

The software automates three-view generation, 3D model creation and rigging before the character is used.

3) Voice processing runs locally

Lip sync and contextual expression switching are processed on the user's PC rather than sending microphone audio to a server for those features.

4) Streaming can be tested before buying points

The bundled character already includes expressions, mouth shapes and an idle motion, allowing users to try the streaming workflow without generating a new avatar.

App Information

For now, the additional important information currently known includes:

• VTuber Original Character Maker officially launched on August 31, 2026
• the application is developed by Sumeragi
• downloads are available for Windows 10/11 64-bit and macOS 14 Sonoma or newer
• the application download and core streaming functions are free
• one ready-made 3D character is included for testing the streaming features
• generating an original character uses paid points
• one 3D character generation costs 800 pt
• one expression set with six mouth shapes costs 200 pt
• background generation costs 60 pt
• video generation costs 85 pt for every second
• generating an input image from a text description costs 25 pt
• the Standard Pack provides 2,000 pt for ¥2,080 including tax
• a 600-point package costs ¥630
• the Plus Pack provides 2,850 pt for ¥2,800
• points expire 180 days after they are granted
• a full-body character image with the arms separated from the torso is recommended
• the input image background can be removed automatically
• the application generates front, side and back views before building the 3D model
• rigging is performed automatically
• expressions can be selected from templates or requested with free-text prompts
• six mouth shapes are generated for lip-sync use
• microphone input drives lip sync in real time
• contextual expression switching uses on-device speech recognition and AI
• microphone audio and derived speech-recognition text used by those features are not sent to a server
• approximately 1.5 GB of AI data is downloaded on first launch
• the official requirements call for at least 5 GB of available storage
• an internet connection is required for AI generation
• a microphone is required for lip sync and voice-based expression features
• OBS can receive the clean stream view through Browser Source without an additional plugin
• generated characters may be used commercially or non-commercially under the Terms of Service
• generated 3D models can be exported, sold or distributed under the stated conditions
• users must hold the appropriate rights to the source artwork
• the developer notes that generated designs are not exclusive and may resemble outputs generated for other users
• the Steam page is available but the software remains unreleased through Steam at the time checked
• Steam lists Japanese, English and Simplified Chinese interface support
• Steam also discloses a connection to the third-party Tripo service for AI content generation
• official information and downloads are available through the VTuber Original Character Maker Official Website
• official updates are available through VTuber Original Character Maker Official X
• the launch announcement is available through the Official X Post
• the trailer is available through the VTuber Original Character Maker Official Trailer
• the store page is available through VTuber Original Character Maker on Steam

Closing

VTuber Original Character Maker attempts to reduce the number of technical steps between owning a character illustration and appearing on stream. One image can become a rigged 3D model with expressions and lip sync before being routed directly into an OBS workflow from the same application.

The Windows and macOS versions are already available through the official website, while the Steam release is still pending. Users also need to account for the paid generation system and ensure they hold the appropriate rights to any illustration used as source material.