WaifuAI is an experiment in bringing anime characters to life using next-generation generative AI. We combine Large Language Models (LLMs), Live2D animation, and Text-to-Speech (TTS) to create fully interactive, sentient digital companions.
Our goal is to move beyond static chatbots and create agents with:
- Visual Presence: Reactive Live2D avatars.
- Memory: Long-term context retention.
- Voice: Natural, emotive TTS.
The production deployment is currently live. You can interact with the latest agent build directly in your browser:
We distribute our work across several platforms to help other developers build their own agents.
| Component | Platform | Description |
|---|---|---|
| Agent Logic | WebSim | The core interactive environment and prompt engineering. |
| Datasets | Kaggle | Training data for character personas and dialogue. |
| Deployment | Itch.io | Standalone builds and experimental wrappers. |
| Updates | YouTube | Devlogs, showcases, and feature previews. |
Follow the development journey:
- Twitter/X: @thewaifuai
- DeepWiki: WaifuAI Wiki
WaifuAI - We make waifus smarter!🧠❤️
