Why We Chose Vector Databases

Traditional keyword search (BM25) is great for exact matches, but it fails at understanding intent. That is why Pluro is built on top of a high-performance vector database.

High-Dimensional Space

We convert your text into vectors—lists of numbers representing semantic meaning. In this high-dimensional space, "dog" and "puppy" are close together, even though they share no letters. This allows us to retrieve memories based on meaning, not just keywords.

This architecture allows Pluro to scale to millions of memories while maintaining millisecond-latency retrieval times.