Inno Agent
A self-hosted AI learning agent with editable learner memory, a study wiki and scheduled review, supporting Ollama alongside cloud model providers.
Alternative to:

Inno Agent is a personal AI learning agent for ongoing tutoring and study. It keeps learning goals, misconceptions and study materials available across conversations, with records you can inspect and correct.
- Learner memory: An editable profile tracks goals, understanding and preferences, while conversation recall brings earlier discussions into later sessions.
- Study wiki: PDF, Office, image and text materials become maintainable pages with source references, concepts and analysis.
- Scheduled review: The agent can schedule spaced review, daily reflection and knowledge-base updates, with reminders through Feishu, WeChat or QQ.
- Programming practice: Practice Lab combines generated exercises with a browser terminal and saved run output, so follow-up explanations can use actual errors and results.
- Teaching and mathematics: Skills and workspace presets support essay feedback, lesson planning and interactive geometry; VeryMath workspaces add mathematical computing, paper study and Lean 4 proof assistance.
It runs as a macOS or Windows desktop app, a self-hosted web interface or a terminal app. Data stays in local files by default. Ollama and local models are supported alongside OpenAI, Anthropic and DeepSeek through compatible endpoints; choosing a cloud provider sends model requests to that service.
Inno Agent is open source under the MIT license. It's designed for one learner per instance and has no multi-user authentication or tenant isolation.