Epsilla is an open-source vector database. Our focus is on ensuring scalability, high performance, and cost-effectiveness of vector search. EpsillaDB bridges the gap between information retrieval and memory retention in Large Language Models.

The key features of Epsilla include:

• High performance and production-scale similarity search for embedding vectors.

• Full fledged database management system with familiar database, table, and field concepts. Vector is just another field type.

• Native Python support and REST API interface.

Epsilla’s core is written in C++ and leverages the advanced academic parallel graph traversal techniques for vector indexing, achieving 10 times faster vector search than HNSW while maintaining precision levels of over 99.9%.

Quick Start

1. Run Backend in Docker

docker pull epsilla/vectordb
docker run --pull=always -d -p 8888:8888 -v /tmp:/tmp epsilla/vectordb

2. Interact with Python Client

from pyepsilla import vectordb

client = vectordb.Client(host='localhost', port='8888')
client.load_db(db_name="MyDB", db_path="/tmp/epsilla")
client.use_db(db_name="MyDB")

client.create_table(
  table_name="MyTable",
  table_fields=[
    {"name": "ID", "dataType": "INT"},
    {"name": "Doc", "dataType": "STRING"},
    {"name": "Embedding", "dataType": "VECTOR_FLOAT", "dimensions": 4}
  ]
)

client.insert(
  table_name="MyTable",
  records=[
    {"ID": 1, "Doc": "Berlin", "Embedding": [0.05, 0.61, 0.76, 0.74]},
    {"ID": 2, "Doc": "London", "Embedding": [0.19, 0.81, 0.75, 0.11]},
    {"ID": 3, "Doc": "Moscow", "Embedding": [0.36, 0.55, 0.47, 0.94