LLM Service

LLM Overview

The LLM (Large Language Model) Service in our real estate platform is designed to interpret user queries expressed in natural language and automatically set the appropriate search filters based on these inputs. This service enhances the user experience by allowing more intuitive search capabilities where users can simply describe what they are looking for in plain language, and the system intelligently translates this into actionable search queries.

How It Works

1. Query Reception: The service receives a user's natural language query via a FastAPI endpoint.
2. Query Parsing: It employs a sophisticated parsing mechanism to interpret the natural language input and translate it into specific search filters.
3. API Schema Generation: The service dynamically generates an OpenAPI schema that represents these filters.
4. Search Execution: Utilizing the parsed filters, the service interacts with the backend to fetch and return the relevant real estate listings that meet the criteria specified by the user's original query.

Key Technologies

For an in-depth look at the technologies used in our LLM service, visit the LLM Technologies page.

Setting Up the LLM Service

To get started with setting up and running the LLM service, refer to the How to Start the Service page.