ai-real-estate-assistant
Open-source AI real estate assistant: RAG + vector search over your own listings, 16 LLM providers, FastAPI + Next.js. Run it on your own server with Docker, or
What is ai-real-estate-assistant?
Open-source AI real estate assistant: RAG + vector search over your own listings, 16 LLM providers, FastAPI + Next.js. Run it on your own server with Docker, or try the live demo.
Overview of ai-real-estate-assistant
What it is
AI Real Estate Assistant is an open-source conversational assistant for property search, financial analysis and market insight. It uses RAG (Retrieval-Augmented Generation) and vector search over your own property listings. The project is MIT licensed and can be deployed on your own server with Docker or on a hosted platform. It supports 17 LLM providers and nine languages. The repository is available at https://github.com/AleksNeStu/ai-real-estate-assistant and a live demo is hosted at https://realty.kittray.com.
Key features
- Natural-language search: ask for a property in plain language and get matched listings from a semantic index.
- Toolbelt behind the chat: mortgage calculator, rent-vs-buy comparison, ROI, TCO, valuation with multi-year price forecast.
- Interactive maps with clustered markers and area analytics.
- Nine languages: English, Polish, Russian, German, Spanish, Italian, Portuguese, Turkish, Ukrainian.
- Supports 17 LLM providers (the README mentions 16 in the tagline and 17 in the body).
- Can be deployed on your own VPS with Docker, on a hosted platform with one click, or using published images.
Use cases
- Property search for real estate agents and buyers using natural language queries.
- Financial analysis of properties including mortgage, ROI, and rent-vs-buy comparisons.
- Market insight and city-level analytics for real estate professionals.
- Comparative market analysis (CMA) wizard for property valuation.
- Document upload and query for knowledge base management.
Getting started
bashgit clone https://github.com/AleksNeStu/ai-real-estate-assistant.gitcd ai-real-estate-assistant# Docker, demo mode, no API keys requiredcp deploy/compose/.env.example deploy/compose/.envdocker compose -f deploy/compose/docker-compose.yml up --build# Frontend: http://localhost:3082 · API: http://localhost:8082/docsOn Windows, use .scripts/docker/start.ps1. For manual setup with uv + npm, see the repository README.
License and pricing
The project is licensed under MIT. It is open-source and free to use. The hosted version (paid) is being built separately.
Based on the project's public README and repository metadata.