Lobe Vidol is a platform for creating and interacting with virtual idols, targeting users interested in virtual companionship and content creation. It offers a cohesive experience with features like text and video chat, character customization, and MMD dance integration, aiming to make virtual idol creation accessible.

How It Works

Lobe Vidol integrates multiple AI model providers (AWS Bedrock, Google AI, Anthropic, etc.) for conversational capabilities, supporting both text and voice (TTS/STT) interactions. It leverages technologies like mmd-parser and three-vrm to enable 3D character animation, including MMD dances and custom character interactions via a character editor. The platform is built as a Progressive Web App (PWA) for a near-native experience across devices.

Quick Start & Requirements

• Online Demo: No installation required, visit the website for a quick experience.
• Local Development: Requires bun for installation and running.
• Dependencies: bun (package manager and runtime).
• Documentation: [Documentation][docs]

Highlighted Details

• Supports a wide array of AI model providers for conversational AI.
• Integrates MMD dance content and character customization with VRM model support.
• Features a marketplace for characters and dances.
• Offers both text and video-based interactions with virtual idols.
• Built as a PWA for cross-device accessibility.

Maintenance & Community

• Active development with open beta testing.
• Community support via Discord.
• Project health indicated by numerous GitHub shields for releases, tests, and community engagement.

Licensing & Compatibility

• Licensed under Apache 2.0.
• Permissive license suitable for commercial use and integration into closed-source projects.

Limitations & Caveats

The project is currently in early development (beta testing), indicating potential for instability or breaking changes. While extensive, the list of supported AI model providers may not include all niche or emerging options.