vrchat-ai-agent
A VRChat AI agent combining speech recognition, local LLMs, VOICEVOX-style TTS, OSC, YOLO/Pose, VLM, memory, emotion models, and autonomous behavior. It controls avatar movement, gaze, expressions, emotes, Use/Grab, and piano. Uses Groq VLM and Gemini Grounding Search or Wikimedia API. Source, prompts, and personal memories are private.
Translated from Japanese · show originalhide original
VRChat向けのAIエージェント実装です。音声認識、ローカルLLM、VOICEVOX系TTS、VRChat OSC、YOLO/Pose、VLM、長期記憶、感情/欲求モデル、自律行動を組み合わせ、会話だけでなくアバターの移動、視線、表情、エモート、Use/Grab、ピアノ演奏まで扱います。視界説明はGroq VLMを優先し、検索が必要な発話ではGemini Grounding SearchとWikimedia APIを使い分けます。ソースコード、キャラクタープロンプト、個人記憶データは公開していません。
- llm
- local-llm
- osc
- python
- voicevox
- vrchat
- vtuber
- whisper
- yolo
- Stars
- 10
- Forks
- 0
- + today
- +2
- Created
- 5mo
Ranking data as of October 5, 2026 (UTC).
Star History
Today, hour by hour
Overview
A VRChat AI agent combining speech recognition, local LLMs, VOICEVOX-style TTS, OSC, YOLO/Pose, VLM, memory, emotion models, and autonomous behavior. It controls avatar movement, gaze, expressions, emotes, Use/Grab, and piano. Uses Groq VLM and Gemini Grounding Search or Wikimedia API. Source, prompts, and personal memories are private. It ranks #407 on GitTiger, gaining +2 stars on October 5, 2026 (UTC).
It was created 5mo ago.
git clone https://github.com/gamio-22/vrchat-ai-agent.git
cd vrchat-ai-agent
# see README for setup