NucliaDB
NucliaDB is a hybrid search database for unstructured data, combining vector, full-text and graph indexes with support for large datasets and multi-tenancy.
NucliaDB is a database for storing and searching unstructured data. It combines vector, full-text and graph indexes in one hybrid search engine, and is built to index large datasets with multi-tenant support.
It stores text, files, vectors, labels and annotations, and supports semantic search over vectors as well as keyword search. Data can be exported in formats used by common NLP pipelines such as Hugging Face datasets and PyTorch. It includes role-based security with upstream proxy authentication, multiple field types (text, file, link, conversation), PostgreSQL as its storage layer, and S3-compatible blob storage with support for GCS and Azure Blob Storage. It also supports replication of index storage and distributed search.
It can optionally connect to the hosted Nuclia cloud for data extraction, enrichment and inference, but works as a standalone database on its own. NucliaDB is written in Rust and Python.