Knowledge Base & Vector Store
Ingest plain text documents, generate 1536-dim embeddings, and query via PostgreSQL pgvector
Active Scope: Demo Developer's Personal (Private to you)
Personal Documents
Upload & Vectorize Knowledge Chunks
Documents are automatically split into ~500 token chunks with 50-token semantic overlap.
File upload disabled in the live demo
Supports plain text files up to 200KB. Automatically chunked at ~500 tokens.
Indexed pgvector Chunks (0)
No documents indexed in this workspace
Upload text files above to enable Contextual RAG search during your AI chats.