Set Up Vector Search

LLM similarity search compares embedding vectors. Momen stores vectors in the relational database for semantic sort and AI context.

Open Data tab → Data model to create tables and enable vector fields.

Database setup: Set Up the Database. AI usage: AI Integration.

Save vector data

1. Data tab → Data model — create a table with a Text field
2. Field menu → Enable Vector Storage — choose an embedding model (add models in Action tab → AI Integration)
3. Text saved in that field is embedded automatically

Vector sorting

Sort by distance between a reference value and each row’s vector — smaller distance = higher similarity.

List semantic search

1. Bind a list to a table with a vector-enabled field
2. Sort → vector field → Vector mode
3. Choose distance method
4. Bind the reference object (e.g. search box text)

AI context (RAG)

1. Create a table with vector-enabled text fields for your knowledge base
2. AI config → Contexts → select table → filter icon
3. Add Sort: vector field, distance method, comparison content bound to user input

Demo: slang search

A case table with internet slang embeddings; search input is vectorized and the list returns nearest matches.

Notes

• Text fields only
• Plan storage for embedded rows
• Pair with Query and Bind Data for UI search

Related Reading

• Set Up the Database
• AI Integration