Open TutorAI
Open TutorAI is an open-source, self-hosted AI tutoring platform built on Open WebUI, offering personalized AI tutors, avatar discussion mode, feedback systems, and role-based permissions for students, teachers, and admins in the classroom.
About this tool
Open TutorAI is an open-source, self-hosted AI tutoring platform built on top of Open WebUI, designed to provide a collaborative AI-powered learning environment for students, teachers, and administrators in the classroom.
Key Capabilities
- Personalized Tutoring: Generate a personalized LLM, optionally paired with a user-selected avatar, tailored to each learner's style, personality, or curriculum.
- Avatar Discussion Mode: Go beyond text with an avatar-powered tutor that integrates voice and video for a face-to-face style of tutoring.
- AI Monitoring Dashboard: Track cognitive load, confusion metrics, and feedback impact.
- Model Builder: Easily create Ollama models, add custom characters/agents, customize chat elements, and import models.
- Image Generation Integration: Incorporate image generation via AUTOMATIC1111 API, ComfyUI (local), or OpenAI DALL-E (external) to enrich chats with dynamic visual content.
- Multilingual Support: Use Open TutorAI in your preferred language through internationalization (i18n).
- Granular Permissions & User Groups: Admins can define detailed roles and permissions for secure, customized user experiences.
- PWA Support: Install on mobile devices for an app-like experience, including offline mode on localhost.
- Docker Setup: Deploy in minutes with :ollama and :cuda tagged images.
Who
It's For Students can join classrooms, build their own tutoring supports, and learn with an avatar tutor. Teachers and admins can manage roles, permissions, and feedback workflows.
Roadmap
- Q2 2025: Initial Release (MVP) — student space, tutor assistant, feedback system, role management
- Q4 2025: Classroom Tools — course creation, assignment workflow, progress reports
- Q1 2026: Multimodal Interaction + Multi-Agent Tutoring System — voice input, webcam data sync, engagement prediction, agent-based architecture
- Mid 2026: Plugin Ecosystem & LLM Integration — plugin system, LLM adapters, pipelines, prompt-tuning
Availability
Open source, hosted on GitHub, with a Discord community for support.