Open-LLM-VTuber is an open-source project that brings AI-powered voice companions to life by connecting LLMs with responsive Live2D avatars.
Open-LLM-VTuber is a modular, cross-platform open-source application designed to run a personalized voice-interactive AI companion locally or in the cloud. It features hands-free conversation with voice interruption support, linking automated speech recognition (ASR) and text-to-speech (TTS) engines directly to a responsive Live2D avatar. The application offers a web mode and a transparent desktop client that functions as a "desktop pet". It supports complex workflows such as long-term memory via Letta and tool execution via the Model Context Protocol (MCP), letting users customize their virtual avatar with a wide variety of local and online AI services.
** Queried the SurrealDB database and analyzed workspace helper scripts (e.g., [test_our_actual_draft.ts](file:///home/bun/.gemini/antigravity-cli/scratch/test_our_actual_draft.ts) and [test_our_actual_draft_acp.ts](file:///home/bun/.gemini/antigravity-cli/scratch/test_our_actual_draft_acp.ts)) to extract formatting requirements, synonym maps, and prior coverage checks.
3. **Web Search:** Conducted a web search to identify the official homepage for `Open-LLM-VTuber` (`https://llmvtuber.com/`) to resolve the URL collapse issue.
4. **Dry Run Testing:** Wrote and executed [test_vtuber_draft.ts](file:///home/bun/.gemini/antigravity-cli/scratch/test_vtuber_draft.ts) to verify the tag drift corrections and URL changes.
5. **Output Generation:** Formatted the audit results using the exact AICrier Markdown report structure.
DISCOVERED
1h ago
2026-06-04
PUBLISHED
1h ago
2026-06-04
RELEVANCE