multi-model database / vector search
UStore is a C++ multi-modal ACID database that stores blobs, documents, graphs, and vectors, exposing Pandas and NetworkX APIs plus C, C++, Python, Java, and Go bindings.
Same problem, same approach. Swapping one for another is a config change, not a rewrite.
pick for a mature native multi-model DB with SQL-like AQL and clustering instead of UStore's C++/Python-first API surface
🥑 ArangoDB is a native multi-model database with flexible data models for documents, graphs, and key-values. Build high performance applications using a convenient SQL-like query language or JavaScript extensions.
pick for a JVM multi-model DB with SQL, Gremlin/TinkerPop and full-text support over a C++-first embedded stack
OrientDB is the most versatile DBMS supporting Graph, Document, Reactive, Full-Text and Geospatial models in one Multi-Model product. OrientDB can run distributed (Multi-Master), supports SQL, ACID Transactions, Full-Text indexing and Reactive Queries.
Solves the same problem with a different architecture or at a different layer. Expect to rewrite the integration.
pick inside .NET when you want an embedded ACID multi-paradigm store with no server process
C# .NET NOSQL ( key value, object store embedded TextSearch SemanticSearch Vector layer ) ACID multi-paradigm database management system.
pick when you want a mature .NET/HTTP document database and can skip graph and native vector search
ACID Document Database
pick for in-memory data structures plus document/vector queries when the dataset fits RAM and persistence is secondary
For developers, who are building real-time data-driven applications, Redis is the preferred, fastest, and most feature-rich cache, data structure server, and document and vector query engine.
pick when you need a dedicated massive-scale vector engine with payload filtering and can drop documents and graphs
Qdrant - High-performance, massive-scale Vector Database and Vector Search Engine for the next generation of AI. Also available in the cloud https://cloud.qdrant.io/
pick for cloud-native distributed vector ANN search at very large scale rather than a bundled multi-modal store
Milvus is a high-performance, cloud-native vector database built for scalable vector ANN search
pick in browser/edge JS for in-process full-text and vector search where you can't ship a server DB
🌌 A complete search engine and RAG pipeline in your browser, server or edge network with support for full-text, vector, and hybrid search in less than 2kb.
pick for real-time SQL-driven full-text/vector search if you accept the GPL-3.0 license
Open-source search database for full-text, vector, and hybrid search with real-time indexing and SQL.
pick when you want a vector DB with GraphQL and structured filtering rather than an embedded C++ multi-modal store
Weaviate is an open-source vector database that stores both objects and vectors, allowing for the combination of vector search with structured filtering with the fault tolerance and scalability of a cloud-native database.
pick for fast typo-tolerant hybrid search behind an HTTP API when transactions and graphs are not needed
A lightning-fast search engine API bringing AI-powered hybrid search to your sites and applications.
pick when you need large-scale ranking and search on a JVM platform and can absorb the heavier ops footprint
The AI search platform
An earlier or more limited way to do the job. Still the right call when you need something small, proven, or CPU-only.
pick if you only need transactional documents and already operate FoundationDB
Distributed, transactional document database backed by FoundationDB.
pick for a minimal embedded Rust document store when you don't need graphs or vectors
simple, embeddable document database with acid guarantees
pick as a raw LSM-backed ACID key-value engine in .NET, below the query and model layer UStore provides
ZoneTree is a persistent, high-performance, transactional, ACID-compliant ordered key-value database and LSM-tree storage engine for .NET.
pick if you're already on ClickHouse for SQL vector search, noting ~19 months without a push
A @ClickHouse fork that supports high-performance vector search and full-text search.
These projects were analysed and named UStore among their alternatives. The relationship is not symmetric — how UStore rates them is a separate judgement, made when UStore is analysed in its own right.
calls UStore “Choose when you need an embeddable multi-modal store with Python/Pandas/NetworkX bindings instead of a Clojure Datalog one.”
A simple, fast and versatile Datalog database
calls UStore “multi-modal database project with no recent maintenance; choose only if you need an old lightweight replacement.”
🥑 ArangoDB is a native multi-model database with flexible data models for documents, graphs, and key-values. Build high performance applications using a convenient SQL-like query language or JavaScript extensions.