A local-first, privacy-preserving personalised AI tutor for students.
LocalAI_V3 is a Windows desktop AI assistant designed for student learning, spoken English practice, document-based RAG, local memory, web-crawled knowledge ingestion, image understanding, and voice interaction.
The project is designed to run locally on the user's own computer, reducing dependence on cloud APIs and keeping personal learning data private.
Key Features
Local GGUF LLM chat
Spoken English practice
Voice input with ASR
Voice output with XTTS
Local RAG over user knowledge files
Web knowledge ingestion into local RAG
Local memory system
Image understanding with Qwen-VL
Streaming live response mode
Kawaii-style PySide6 desktop UI
Windows easy-start scripts
Project Positioning
LocalAI_V3 is not just a chatbot. It is designed as a local AI tutor that can:
listen to students
answer questions
use personal study materials
remember learning context locally
support spoken practice
analyse images and screenshots
ingest web knowledge into a local knowledge base
The long-term goal is to explore how privacy-preserving local AI systems can support personalised learning.
